INDEX

Explanations

abbreviations or acronyms ending in 'SP'

oai_token-act-pair · gpt-3.5-turbo

specific abbreviations or acronyms related to various topics or entities

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

wcs

-0.65

lang

-0.60

taboola

-0.56

 ..."

-0.56

isine

-0.55

 dawn

-0.55

 Riley

-0.54

-0.53

 Herm

-0.53

 Sprite

-0.53

POSITIVE LOGITS

DERR

0.82

LET

0.80

ODUCT

0.80

ERAL

0.78

ATES

0.76

eers

0.74

GROUP

0.74

ICA

0.72

onica

0.70

IUM

0.69

Activations Density 0.158%

abbreviations or acronyms ending in 'SP'

specific abbreviations or acronyms related to various topics or entities

No Comments

No Known Activations