INDEX

Explanations

phrases indicating actions or statements made publicly by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "took to" followed by various platforms or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

eele

-0.69

ONSORED

-0.64

/-

-0.62

 dwarves

-0.62

 nominated

-0.58

<-

-0.56

 discriminated

-0.55

iasis

-0.55

 Primordial

-0.53

plates

-0.53

POSITIVE LOGITS

pless

0.73

pload

0.69

 silence

0.67

ruciating

0.63

lid

0.62

ãĥīãĥ©

0.60

ghai

0.58

 COURT

0.57

unker

0.57

 market

0.57

Activations Density 0.043%

phrases indicating actions or statements made publicly by individuals

instances of the phrase "took to" followed by various platforms or actions

No Comments

No Known Activations