INDEX

Explanations

phrases related to conflict, betrayal, and subjective value

oai_token-act-pair · gpt-3.5-turbo

significant contrasts or shifts in tone within narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 withdrawal

-0.80

 withdraw

-0.79

 sacked

-0.78

hement

-0.77

 pensions

-0.75

abwe

-0.73

illance

-0.73

 exchanged

-0.72

 legally

-0.70

 bloc

-0.70

POSITIVE LOGITS

Favorite

1.17

Advertisement

1.12

Episode

1.04

Anyway

1.00

advertisement

0.95

Fun

0.91

SPONSORED

0.90

Includes

0.89

Recommended

0.89

Related

0.89

Activations Density 0.448%

phrases related to conflict, betrayal, and subjective value

significant contrasts or shifts in tone within narratives

No Comments

No Known Activations