INDEX

Explanations

words related to 'ting' or phrases that include 'ting'

oai_token-act-pair · gpt-3.5-turbo

words related to ongoing actions or processes

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 weap

-0.66

Mub

-0.65

Caf

-0.65

 analys

-0.64

agos

-0.64

 reviewing

-0.64

£ı

-0.64

ici

-0.62

 paramed

-0.62

 medi

-0.62

POSITIVE LOGITS

ting

1.17

ton

0.98

tons

0.96

mouth

0.93

ling

0.91

ted

0.90

ding

0.88

les

0.87

lling

0.82

bing

0.78

Activations Density 0.015%

words related to 'ting' or phrases that include 'ting'

words related to ongoing actions or processes

No Known Activations