INDEX

Explanations

instances where the word "none" is mentioned

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing the concept of "none" or the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

bledon

-0.72

srf

-0.70

illon

-0.67

 widest

-0.64

DA

-0.62

lished

-0.59

urry

-0.59

romy

-0.57

rote

-0.56

gnu

-0.55

POSITIVE LOGITS

conom

0.96

theless

0.85

uther

0.81

xus

0.80

essee

0.79

lust

0.75

except

0.74

galitarian

0.74

 Detected

0.73

 else

0.70

Activations Density 0.017%

instances where the word "none" is mentioned

phrases emphasizing the concept of "none" or the absence of something

No Comments

No Known Activations