INDEX

Explanations

terms related to coding, programming, and technical instructions

oai_token-act-pair · gpt-3.5-turbo

references to structured activities or formal processes

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

 Priebus

-0.61

0000000000000000

-0.59

jri

-0.56

 resignation

-0.56

awatts

-0.54

PAC

-0.54

missing

-0.54

timeout

-0.54

 retweet

-0.54

BTC

-0.53

POSITIVE LOGITS

 contexts

0.86

lance

0.79

 Occupations

0.73

 literature

0.70

 folklore

0.68

 circles

0.67

 Jung

0.66

£ı

0.65

emporary

0.64

Era

0.63

Activations Density 0.869%

terms related to coding, programming, and technical instructions

references to structured activities or formal processes

No Comments

No Known Activations