INDEX

Explanations

pronouns 'I' and 'we'

oai_token-act-pair · gpt-3.5-turbo

instances of the pronoun "I" and references to personal experiences or actions

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

 Constitutional

-0.79

 PRESS

-0.70

 Newspaper

-0.68

 understatement

-0.66

 Gender

-0.66

 supra

-0.64

 Triple

-0.64

berra

-0.63

 Press

-0.60

 irony

-0.60

POSITIVE LOGITS

 could

0.84

ctic

0.84

uler

0.83

'll

0.82

 wouldn

0.77

sylv

0.76

 shouldn

0.75

inki

0.75

ilant

0.75

llular

0.74

Activations Density 0.125%

pronouns 'I' and 'we'

instances of the pronoun "I" and references to personal experiences or actions

No Comments

No Known Activations