INDEX

Explanations

words and phrases related to introducing oneself or presenting information

oai_token-act-pair · gpt-3.5-turbo

references to self-presentation and identity

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

ruce

-0.66

erest

-0.61

sterdam

-0.60

iking

-0.58

 Herb

-0.57

ciples

-0.56

ï¸ı

-0.55

 Tradable

-0.55

pps

-0.54

enfranch

-0.53

POSITIVE LOGITS

ICAN

0.73

 predicament

0.71

ional

0.70

æ³

0.70

thin

0.66

 threat

0.66

 grievances

0.66

 dilemma

0.66

ISC

0.64

ographics

0.63

Activations Density 0.127%

words and phrases related to introducing oneself or presenting information

references to self-presentation and identity

No Comments

No Known Activations