INDEX

Explanations

actions or instructions related to performing tasks or operations

oai_token-act-pair · gpt-3.5-turbo

action verbs and terms related to engagement or interaction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Seym

-0.78

 Beir

-0.68

 submar

-0.68

 Frie

-0.67

 Vaugh

-0.64

 Nare

-0.64

 Maver

-0.63

 Niet

-0.63

 Palestin

-0.62

 Azerb

-0.60

POSITIVE LOGITS

 guiActiveUnfocused

0.78

ments

0.78

ables

0.77

able

0.73

yon

0.72

0.70

backs

0.70

 html

0.69

ername

0.68

Activations Density 0.124%

actions or instructions related to performing tasks or operations

action verbs and terms related to engagement or interaction

No Comments

No Known Activations