INDEX

Explanations

mentions of top rankings or positions in various categories

oai_token-act-pair · gpt-3.5-turbo

entities and classifications related to rankings and categories

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

rocal

-0.86

ref

-0.70

pered

-0.69

ickle

-0.68

rint

-0.68

reads

-0.67

EngineDebug

-0.66

ittle

-0.65

Kal

-0.64

pload

-0.63

POSITIVE LOGITS

 imaginable

0.97

(>

0.95

 EVER

0.91

 Universities

0.83

 ever

0.76

ever

0.69

 Ever

0.68

 tert

0.67

Ever

0.65

viz

0.64

Activations Density 0.322%

mentions of top rankings or positions in various categories

entities and classifications related to rankings and categories

No Comments

No Known Activations