INDEX

Explanations

numerical figures or statistics

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "number" followed by numerical values

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rador

-0.86

Interstitial

-0.76

inar

-0.76

sbm

-0.74

clair

-0.72

osta

-0.67

INTON

-0.66

mosp

-0.66

 Shroud

-0.66

ossier

-0.65

POSITIVE LOGITS

 numbers

0.95

number

0.91

 number

0.89

 Numbers

0.87

 digits

0.85

encies

0.74

plates

0.73

########

0.72

esses

0.72

 bells

0.71

Activations Density 0.040%

numerical figures or statistics

occurrences of the word "number" followed by numerical values

No Comments

No Known Activations