INDEX

Explanations

names and terms that might be difficult to interpret in the English language

oai_token-act-pair · gpt-3.5-turbo

specific performer names or groups associated with a circus-like theme

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ufact

-0.75

artney

-0.65

icipated

-0.64

thood

-0.62

emouth

-0.61

angelo

-0.58

ikuman

-0.58

retty

-0.58

luster

-0.56

etooth

-0.55

POSITIVE LOGITS

 Herald

0.62

×Ļ×

0.60

Pv

0.55

 Poles

0.54

 coli

0.53

mess

0.53

oxide

0.52

å£«

0.52

 Zeit

0.51

Ü

0.50

Activations Density 1.013%

names and terms that might be difficult to interpret in the English language

specific performer names or groups associated with a circus-like theme

No Comments

No Known Activations