INDEX

Explanations

information or pieces of news

oai_token-act-pair · gpt-3.5-turbo

terms related to acquiring information or insights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

idency

-0.66

LV

-0.62

PORT

-0.62

utsche

-0.62

egal

-0.61

ugal

-0.61

wcsstore

-0.60

 youth

-0.59

 seniors

-0.59

 moderators

-0.59

POSITIVE LOGITS

selves

0.97

eous

0.86

hest

0.83

sonian

0.83

ouncing

0.81

mares

0.80

sheet

0.79

arded

0.76

rooms

0.76

iversary

0.75

Activations Density 0.072%

information or pieces of news

terms related to acquiring information or insights

No Comments

No Known Activations