INDEX

Explanations

phrases indicating the start of a new section or information

oai_token-act-pair · gpt-3.5-turbo

the word "Here" in various contexts within the document

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

ONSORED

-0.81

è¦ļéĨĴ

-0.68

ãĥ¯ãĥ³

-0.64

TPPStreamerBot

-0.61

 "$:/

-0.59

Edge

-0.58

 feasibility

-0.58

wcs

-0.57

ãģ®å®

-0.56

ãĥīãĥ©

-0.56

POSITIVE LOGITS

tic

1.11

tics

1.10

's

1.09

are

1.07

tical

1.06

after

0.92

 Comes

0.92

 comes

0.89

're

0.89

 goes

0.87

Activations Density 0.050%

phrases indicating the start of a new section or information

the word "Here" in various contexts within the document

No Comments

No Known Activations