INDEX

Explanations

news-related entities or events mentioned in reports

oai_token-act-pair · gpt-3.5-turbo

phrases related to reporting and attribution of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

istries

-0.88

rontal

-0.87

aden

-0.76

etheless

-0.74

uga

-0.72

Cola

-0.72

teness

-0.69

joice

-0.67

cking

-0.67

incial

-0.66

POSITIVE LOGITS

by

0.92

 herein

0.84

 above

0.83

 evidenced

0.80

 previously

0.78

 below

0.77

 exempl

0.70

BY

0.69

 Kurd

0.69

 happens

0.67

Activations Density 0.063%

news-related entities or events mentioned in reports

phrases related to reporting and attribution of information

No Comments

No Known Activations