INDEX

Explanations

phrases related to having limited choices or options

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ahime

-0.75

ivas

-0.73

hn

-0.61

avia

-0.61

gart

-0.61

RAFT

-0.59

 Elements

-0.58

otype

-0.57

sea

-0.56

rev

-0.56

POSITIVE LOGITS

 whatsoever

0.86

except

0.86

 except

0.83

 besides

0.78

nor

0.75

nor

0.68

 resembling

0.68

 wrong

0.68

imilar

0.67

soever

0.64

Activations Density 0.181%

phrases related to having limited choices or options

No Comments

No Known Activations