INDEX

Explanations

first-person statements indicating action or sharing information

oai_token-act-pair · gpt-3.5-turbo

phrases that include the speaker's perspective or direct address

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ELD

-0.70

ILD

-0.68

NAS

-0.64

bidden

-0.59

 Associated

-0.59

 inaction

-0.58

ibles

-0.58

required

-0.58

gradient

-0.58

senal

-0.58

POSITIVE LOGITS

 assume

0.99

 assure

0.93

 emphasize

0.93

 clarify

0.92

 subscribe

0.90

 give

0.89

 pray

0.88

 suppose

0.87

 proceed

0.86

 guess

0.84

Activations Density 0.040%

first-person statements indicating action or sharing information

phrases that include the speaker's perspective or direct address

No Comments

No Known Activations