INDEX

Explanations

descriptive words related to articles and written content

oai_token-act-pair · gpt-3.5-turbo

references to original content or articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

cellaneous

-0.69

ileged

-0.68

ensitive

-0.68

rily

-0.67

lov

-0.66

minist

-0.66

quished

-0.64

lain

-0.63

eric

-0.63

enged

-0.62

POSITIVE LOGITS

ãĤ¨ãĥ«

0.78

SpaceEngineers

0.73

ãĥ¼ãĤ¯

0.70

 guiActiveUnfocused

0.69

appa

0.67

ãĥĬ

0.66

ãĤ·ãĥ£

0.66

ãĤ¿

0.63

ItemImage

0.63

æī

0.61

Activations Density 0.143%

descriptive words related to articles and written content

references to original content or articles

No Comments

No Known Activations