INDEX

Explanations

proper nouns and dates, possibly related to legal or historical contexts

oai_token-act-pair · gpt-3.5-turbo

special characters or indicators of text boundaries

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

bg

-0.76

leeve

-0.72

xxx

-0.72

antes

-0.71

agi

-0.69

zac

-0.67

arettes

-0.67

eno

-0.67

thood

-0.67

acy

-0.66

POSITIVE LOGITS

 largest

1.06

 biggest

1.01

 majority

0.97

resa

0.96

 youngest

0.95

 heaviest

0.90

 oldest

0.90

 vast

0.90

 latter

0.89

 longest

0.89

Activations Density 0.383%

proper nouns and dates, possibly related to legal or historical contexts

special characters or indicators of text boundaries

No Comments

No Known Activations