INDEX

Explanations

statements that refer to a narrative or a sequence of events

oai_token-act-pair · gpt-3.5-turbo

mentions of "story."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ignt

-0.77

orem

-0.76

emale

-0.73

aez

-0.71

inence

-0.69

ynski

-0.68

anyon

-0.65

ardless

-0.62

ategory

-0.61

 numeric

-0.61

POSITIVE LOGITS

telling

1.33

te

1.13

boards

1.01

book

1.00

tell

0.98

arc

0.97

board

0.94

 arcs

0.90

books

0.89

boarding

0.86

Activations Density 0.086%

statements that refer to a narrative or a sequence of events

mentions of "story."

No Comments

No Known Activations