INDEX

Explanations

medical terms from various fields, such as anatomy, physiology, and biology

oai_token-act-pair · gpt-3.5-turbo

suffixes related to specific themes or motifs, particularly those indicating characterization or cultural references

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enhagen

-0.98

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.81

PDATE

-0.70

avorite

-0.68

 skelet

-0.67

omnia

-0.66

isSpecialOrderable

-0.66

liners

-0.65

PsyNetMessage

-0.64

 prosecut

-0.63

POSITIVE LOGITS

×Ļ

0.77

Õ

0.76

å§

0.76

acter

0.72

iken

0.70

lain

0.70

anooga

0.69

obos

0.68

leon

0.68

Ñı

0.68

Activations Density 0.232%

medical terms from various fields, such as anatomy, physiology, and biology

suffixes related to specific themes or motifs, particularly those indicating characterization or cultural references

No Comments

No Known Activations