INDEX

Explanations

phrases indicating uncertainty or indecision

oai_token-act-pair · gpt-3.5-turbo

references to indecision or uncertainty about actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

arity

-0.81

////////////////

-0.60

 Chap

-0.60

hyde

-0.60

quad

-0.59

////////////////////////////////

-0.58

mony

-0.58

 Barrett

-0.57

 thwarted

-0.57

ĵ

-0.55

POSITIVE LOGITS

 expect

1.38

 Expect

0.98

do

0.92

 look

0.89

eat

0.88

 wear

0.86

 prioritize

0.86

 anticipate

0.83

ask

0.81

 believe

0.81

Activations Density 0.037%

phrases indicating uncertainty or indecision

references to indecision or uncertainty about actions

No Comments

No Known Activations