INDEX

Explanations

phrases related to maintaining anonymity or staying in a particular state

oai_token-act-pair · gpt-3.5-turbo

the word "remain" and its variations in various contexts

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

ramid

-0.86

ongyang

-0.75

isson

-0.70

 Relief

-0.67

 Tribune

-0.65

rote

-0.64

DIT

-0.64

insula

-0.63

oglobin

-0.63

onz

-0.62

POSITIVE LOGITS

 unchanged

1.13

 intact

0.94

 afloat

0.93

 undecided

0.88

 unaffected

0.84

 unanswered

0.83

 undet

0.82

 undefeated

0.81

 untouched

0.81

 silent

0.79

Activations Density 0.030%

phrases related to maintaining anonymity or staying in a particular state

the word "remain" and its variations in various contexts

No Comments

No Known Activations