INDEX

Explanations

references to job offers

oai_token-act-pair · gpt-3.5-turbo

terms related to job offers and proposals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

cling

-0.72

Cav

-0.69

Geo

-0.68

 Frankfurt

-0.65

ansk

-0.65

 Cabin

-0.62

 destruct

-0.60

english

-0.60

icent

-0.60

 Headquarters

-0.60

POSITIVE LOGITS

holder

0.81

ļéĨĴ

0.76

vier

0.75

 offered

0.73

VERTISEMENT

0.73

holders

0.72

backs

0.72

 offer

0.71

 offers

0.70

offer

0.70

Activations Density 0.021%

references to job offers

terms related to job offers and proposals

No Comments

No Known Activations