INDEX

Explanations

words related to obstacles or physical blockages

oai_token-act-pair · gpt-3.5-turbo

instances of the terms related to obstruction or prevention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

gow

-0.86

ller

-0.83

llers

-0.80

EMBER

-0.77

brates

-0.67

brate

-0.67

lli

-0.65

mble

-0.63

enture

-0.62

gom

-0.62

POSITIVE LOGITS

 blocking

0.91

quote

0.82

buster

0.78

 blockade

0.78

 obstruct

0.76

busters

0.75

osuke

0.74

aded

0.74

chains

0.73

picking

0.69

Activations Density 0.017%

words related to obstacles or physical blockages

instances of the terms related to obstruction or prevention

No Comments

No Known Activations