INDEX

Explanations

specific verbs indicating author's argument or direction in a text

oai_token-act-pair · gpt-3.5-turbo

phrases that suggest implications or conclusions based on evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ovo

-0.60

nik

-0.58

 herself

-0.58

 Tycoon

-0.57

ledged

-0.56

lance

-0.56

legate

-0.55

osa

-0.55

culus

-0.54

llan

-0.54

POSITIVE LOGITS

 abound

0.71

 pertaining

0.67

 differ

0.65

 respective

0.65

including

0.63

 including

0.63

 Rohing

0.63

 backgrounds

0.62

 arise

0.61

 timelines

0.59

Activations Density 1.518%

specific verbs indicating author's argument or direction in a text

phrases that suggest implications or conclusions based on evidence

No Comments

No Known Activations