INDEX

Explanations

mentions of missions or goals

oai_token-act-pair · gpt-3.5-turbo

references to missions and goals

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

ãģį

-0.84

picking

-0.82

Column

-0.77

 Ange

-0.70

urrency

-0.68

urrencies

-0.68

cling

-0.66

batch

-0.66

amera

-0.66

ª

-0.64

POSITIVE LOGITS

eers

0.90

 mission

0.89

arily

0.88

 missions

0.84

 creep

0.81

 Impossible

0.81

Vie

0.76

 accomplished

0.75

 abroad

0.74

 Mission

0.73

Activations Density 0.019%

mentions of missions or goals

references to missions and goals

No Comments

No Known Activations