INDEX

Explanations

short phrases or keywords related to decision-making or options

oai_token-act-pair · gpt-3.5-turbo

various forms of self-reflection and introspection

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

Rated

-0.68

verend

-0.68

ieve

-0.64

ãģ®éŃĶ

-0.64

ario

-0.63

EH

-0.63

ume

-0.62

currency

-0.61

 Bless

-0.58

Connector

-0.58

POSITIVE LOGITS

 underest

0.94

 somew

0.86

 typo

0.84

 underestimated

0.83

rosso

0.80

 overest

0.80

 someday

0.79

 misunder

0.76

itors

0.76

 harb

0.74

Activations Density 0.578%

short phrases or keywords related to decision-making or options

various forms of self-reflection and introspection

No Comments

No Known Activations