INDEX

Explanations

comparisons or preferences between two options, typically favoring one over the other

oai_token-act-pair · gpt-3.5-turbo

comparisons that emphasize preference for one option over another

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

minent

-0.75

amba

-0.75

mberg

-0.75

adium

-0.75

eria

-0.74

ruary

-0.73

ppo

-0.73

cision

-0.72

endale

-0.71

elaide

-0.70

POSITIVE LOGITS

 than

0.78

 preferring

0.72

 unimagin

0.72

 pricey

0.70

 trivial

0.68

Ide

0.68

 inconvenient

0.66

 innocuous

0.66

 unpleasant

0.65

Leh

0.65

Activations Density 0.015%

comparisons or preferences between two options, typically favoring one over the other

comparisons that emphasize preference for one option over another

No Comments

No Known Activations