INDEX

Explanations

phrases related to removal or extraction

oai_token-act-pair · gpt-3.5-turbo

phrases indicating actions of taking something out or removing it from a context

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

ndra

-0.81

orth

-0.73

thumbnails

-0.72

tions

-0.72

ulty

-0.72

itious

-0.69

-+-+

-0.66

appa

-0.66

margin

-0.65

Origin

-0.65

POSITIVE LOGITS

 seriously

0.94

 Seriously

0.80

 lightly

0.76

cue

0.73

 tumble

0.70

 stride

0.70

 hostage

0.68

 aback

0.67

lim

0.66

 Lenin

0.66

Activations Density 0.243%

phrases related to removal or extraction

phrases indicating actions of taking something out or removing it from a context

No Comments

No Known Activations