INDEX

Explanations

mentions of specific nouns related to various topics such as art, movies, politics, sports, and technology

oai_token-act-pair · gpt-3.5-turbo

terms related to art and its various interpretations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ittees

-0.65

âĵĺ

-0.64

éŃĶ

-0.62

ãĥĥãĥĪ

-0.61

Tos

-0.60

ospital

-0.60

antha

-0.59

DEN

-0.59

 Flavoring

-0.59

issan

-0.59

POSITIVE LOGITS

 sucks

0.90

ain

0.87

!'

0.84

 piss

0.81

!"

0.81

 anymore

0.80

 itself

0.80

manship

0.77

 hates

0.76

 hating

0.76

Activations Density 0.613%

mentions of specific nouns related to various topics such as art, movies, politics, sports, and technology

terms related to art and its various interpretations

No Comments

No Known Activations