INDEX

Explanations

updated events or information

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Updated" and related terms indicating status changes or updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

teenth

-0.82

esm

-0.75

marked

-0.64

cipled

-0.63

LOD

-0.60

 Duff

-0.60

OH

-0.60

jp

-0.59

Redd

-0.59

etc

-0.59

POSITIVE LOGITS

 Date

0.65

 Chronicle

0.61

irement

0.60

 Transcript

0.60

 Coverage

0.58

 Dates

0.58

 Explan

0.57

Updated

0.57

 Content

0.56

escription

0.56

Activations Density 0.015%

updated events or information

instances of the word "Updated" and related terms indicating status changes or updates

No Comments

No Known Activations