INDEX

Explanations

references to small amounts or incremental changes

oai_token-act-pair · gpt-3.5-turbo

expressions indicating small quantities or increments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

aring

-0.78

DragonMagazine

-0.72

Ò

-0.71

urated

-0.70

 Primordial

-0.68

Built

-0.66

SPONSORED

-0.65

psons

-0.64

kefeller

-0.64

ukes

-0.64

POSITIVE LOGITS

 stra

0.86

 longer

0.86

bit

0.82

 differently

0.81

 taller

0.76

 slower

0.76

 sooner

0.74

 faster

0.74

stairs

0.71

 consecut

0.71

Activations Density 0.123%

references to small amounts or incremental changes

expressions indicating small quantities or increments

No Comments

No Known Activations