INDEX

Explanations

terms or phrases related to specific concepts or phenomena

oai_token-act-pair · gpt-3.5-turbo

terms related to scientific phenomena and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

icipated

-0.74

 muted

-0.65

 weap

-0.64

 truthful

-0.62

 hurd

-0.61

etheless

-0.61

 lavish

-0.60

laim

-0.59

 ravaged

-0.59

ictions

-0.59

POSITIVE LOGITS

"—

0.73

".

0.71

 Versus

0.69

 Curve

0.68

 Syndrome

0.68

 versus

0.66

 Knot

0.66

Geh

0.65

",

0.64

 Route

0.64

Activations Density 0.446%

terms or phrases related to specific concepts or phenomena

terms related to scientific phenomena and concepts

No Comments

No Known Activations