INDEX

Explanations

words related to specific names and events

oai_token-act-pair · gpt-3.5-turbo

specific character sequences or patterns in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

hett

-0.65

artifacts

-0.65

CoC

-0.62

etsk

-0.62

ELL

-0.61

ulhu

-0.60

audi

-0.60

ashtra

-0.60

lasses

-0.59

furt

-0.59

POSITIVE LOGITS

uala

0.83

 Alvarez

0.70

jah

0.67

 Camer

0.64

 Sharif

0.63

 Strauss

0.62

oglu

0.62

eming

0.61

 Doctrine

0.59

 phosphate

0.59

Activations Density 0.215%

words related to specific names and events

specific character sequences or patterns in text

No Comments

No Known Activations