INDEX

Explanations

unintelligible characters

oai_token-act-pair · gpt-3.5-turbo

uncommon or special characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

RAW

-0.72

ATURE

-0.70

atos

-0.64

 breeze

-0.64

Wan

-0.62

SpaceEngineers

-0.61

FORMATION

-0.59

 Planning

-0.59

 Dragonbound

-0.57

 antioxid

-0.57

POSITIVE LOGITS

iliate

0.82

tg

0.80

raltar

0.78

Footnote

0.70

gp

0.68

enfranch

0.68

ophon

0.66

Corp

0.66

glers

0.66

ilaterally

0.65

Activations Density 0.000%

unintelligible characters

uncommon or special characters

No Comments

No Known Activations