INDEX

Explanations

detailed descriptions or explanations of events or situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "detail" in various forms

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

 Sabha

-0.78

lahoma

-0.68

Enlarge

-0.68

heric

-0.67

ometer

-0.65

ansas

-0.62

ernels

-0.62

ernel

-0.62

itsch

-0.61

ellig

-0.61

POSITIVE LOGITS

 excerpts

0.83

Ĥª

0.83

 details

0.82

stories

0.80

how

0.78

 anecdotes

0.77

 tales

0.77

 feats

0.74

 detailing

0.74

 Charges

0.73

Activations Density 0.093%

detailed descriptions or explanations of events or situations

instances of the word "detail" in various forms

No Comments

No Known Activations