INDEX

Explanations

words or phrases that are commonly used

oai_token-act-pair · gpt-3.5-turbo

the frequency of the word "commonly" in various contexts

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

anmar

-0.72

gur

-0.71

andra

-0.70

agate

-0.70

arta

-0.69

 Fury

-0.69

utenberg

-0.67

ilion

-0.67

ongyang

-0.65

acity

-0.65

POSITIVE LOGITS

 abbrevi

0.99

 referred

0.89

 identifiable

0.89

 categorized

0.88

 regarded

0.88

 encountered

0.88

 practiced

0.85

 seen

0.85

entimes

0.84

ensical

0.83

Activations Density 0.023%

words or phrases that are commonly used

the frequency of the word "commonly" in various contexts

No Comments

No Known Activations