INDEX

Explanations

numerical values indicating quantity or amount

oai_token-act-pair · gpt-3.5-turbo

quantifiable data related to casualties or fatalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tions

-0.84

taboola

-0.81

SourceFile

-0.73

ãĥ¼ãĤ¯

-0.68

Reviewer

-0.67

eries

-0.66

ohydrate

-0.65

axter

-0.65

onomic

-0.64

cum

-0.62

POSITIVE LOGITS

 Beir

0.72

 eight

0.71

 nine

0.71

eight

0.70

six

0.70

 Trayvon

0.69

 four

0.69

six

0.69

 five

0.67

Nin

0.67

Activations Density 0.018%

numerical values indicating quantity or amount

quantifiable data related to casualties or fatalities

No Comments

No Known Activations