INDEX

Explanations

clickable text prompts

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "Click" indicating interactions or actions related to links or buttons

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

utive

-0.80

 Scotia

-0.73

 Chancellor

-0.68

 Croat

-0.65

ASC

-0.62

VAT

-0.61

val

-0.60

OUN

-0.59

 Warsaw

-0.59

 tracts

-0.57

POSITIVE LOGITS

prints

0.81

Click

0.80

click

0.78

Sign

0.76

lish

0.75

whe

0.71

er

0.71

esome

0.70

ety

0.70

tips

0.70

Activations Density 0.013%

clickable text prompts

phrases that include the word "Click" indicating interactions or actions related to links or buttons

No Comments

No Known Activations