INDEX

Explanations

words or phrases related to negation or inability

oai_token-act-pair · gpt-3.5-turbo

the negative contraction "can't"

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

CVE

-0.84

 behavi

-0.66

 SetTextColor

-0.66

 divides

-0.62

interstitial

-0.61

 Gleaming

-0.61

 halves

-0.60

 ancest

-0.59

 tremend

-0.59

çĦ

-0.59

POSITIVE LOGITS

urtle

1.02

weet

0.93

otally

0.88

ruly

0.86

Í

0.86

reprene

0.84

omp

0.81

imately

0.81

acters

0.81

abet

0.80

Activations Density 0.033%

words or phrases related to negation or inability

the negative contraction "can't"

No Comments

No Known Activations