INDEX

Explanations

proper nouns related to politics and business

oai_token-act-pair · gpt-3.5-turbo

names of political figures and institutions

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

Not in Any Lists

No Comments

Negative Logits

aries

-0.68

 Scotia

-0.65

 Giuliani

-0.63

erm

-0.63

arding

-0.62

 Ukrainians

-0.61

 Kern

-0.60

 Pascal

-0.60

 Matthews

-0.60

arded

-0.58

POSITIVE LOGITS

pin

0.85

scribe

0.80

Ļ½

0.79

manship

0.79

é¾įåĸļå£«

0.78

mare

0.75

aye

0.75

daq

0.74

elia

0.72

metic

0.72

Activations Density 0.024%

proper nouns related to politics and business

names of political figures and institutions

No Comments

No Known Activations