INDEX

Explanations

phrases related to actions or intentions of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their actions or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

BDS

-0.65

endif

-0.63

lif

-0.60

cens

-0.60

 Passion

-0.59

ardon

-0.58

lov

-0.57

venge

-0.56

 Bless

-0.56

 inhibition

-0.56

POSITIVE LOGITS

 debuted

0.89

 realised

0.89

 came

0.89

 ventured

0.88

 confronted

0.87

 arrived

0.87

 unveiled

0.84

 entered

0.84

saw

0.82

 joined

0.82

Activations Density 0.126%

phrases related to actions or intentions of individuals

references to individuals and their actions or characteristics

No Comments

No Known Activations