INDEX

Explanations

technical terms related to system capability and functionality

oai_token-act-pair · gpt-3.5-turbo

empty tokens or null values in the text

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

farious

-0.71

 challeng

-0.68

enegger

-0.67

Interstitial

-0.67

 Niet

-0.65

 Vaugh

-0.65

abee

-0.64

agra

-0.61

 withd

-0.59

Ire

-0.59

POSITIVE LOGITS

 unfocusedRange

0.64

lessly

0.60

acan

0.56

 partName

0.55

Â¶

0.54

 pumpkin

0.53

 largeDownload

0.50

Lov

0.50

Activations Density 0.214%

technical terms related to system capability and functionality

empty tokens or null values in the text

No Comments

No Known Activations