INDEX

Explanations

proper nouns (names of people, organizations, events) related to specific actions that they have taken

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "took."

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

 agre

-0.69

 david

-0.66

eers

-0.65

Cong

-0.65

 Smile

-0.65

 sidebar

-0.64

ler

-0.63

earances

-0.63

holm

-0.62

 livest

-0.61

POSITIVE LOGITS

 advantage

1.01

aways

0.97

 heed

0.93

 aback

0.91

autions

0.85

 care

0.84

 precedence

0.84

arnaev

0.80

prising

0.80

 refuge

0.79

Activations Density 0.107%

proper nouns (names of people, organizations, events) related to specific actions that they have taken

occurrences of the word "took."

No Comments

No Known Activations