INDEX

Explanations

words related to limitations, exclusions, and restrictions

oai_token-act-pair · gpt-3.5-turbo

terms related to restrictions and exclusions

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

estyles

-0.60

ebus

-0.59

WM

-0.58

sonian

-0.58

 Thumbnails

-0.56

 Odyssey

-0.55

TIME

-0.55

Bench

-0.55

 Fired

-0.54

oÄŁ

-0.54

POSITIVE LOGITS

 from

1.28

 FROM

1.06

from

1.02

 altogether

0.94

 From

0.90

From

0.85

 entirely

0.83

minded

0.82

iates

0.74

 lest

0.72

Activations Density 0.152%

words related to limitations, exclusions, and restrictions

terms related to restrictions and exclusions

No Comments

No Known Activations