INDEX

Explanations

technical or specific jargon related to various subjects

oai_token-act-pair · gpt-3.5-turbo

terms related to legal and governmental structures or concepts

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

 ..."

-0.69

 thereto

-0.61

 âĢ¦"

-0.60

 thereof

-0.58

..."

-0.56

 respectively

-0.56

...

-0.54

Interstitial

-0.54

](

-0.53

 è£ı

-0.52

POSITIVE LOGITS

 Profile

0.62

 Advice

0.59

 Analysis

0.57

xtap

0.56

rimination

0.56

 Explan

0.56

 References

0.55

 Appeal

0.53

 Benefits

0.53

 Differences

0.53

Activations Density 0.413%

technical or specific jargon related to various subjects

terms related to legal and governmental structures or concepts

No Comments

No Known Activations