INDEX

Explanations

phrases indicating restrictions or permissions involving the sharing or distribution of content

oai_token-act-pair · gpt-3.5-turbo

terms related to the distribution and publication of content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ties

-0.84

 haunt

-0.72

gone

-0.66

 haun

-0.64

outer

-0.62

teenth

-0.61

 colonization

-0.60

tty

-0.59

lif

-0.59

 submar

-0.56

POSITIVE LOGITS

ONSORED

0.70

 herein

0.65

 Mechdragon

0.63

âĢ¦)

0.62

.)

0.62

atform

0.61

 ILCS

0.60

UTH

0.59

 widely

0.59

 freely

0.58

Activations Density 0.020%

phrases indicating restrictions or permissions involving the sharing or distribution of content

terms related to the distribution and publication of content

No Comments

No Known Activations