INDEX

Explanations

locations or settings

oai_token-act-pair · gpt-3.5-turbo

the word "as" in various contexts

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

itiveness

-0.66

eeee

-0.65

ATIVE

-0.62

Constructed

-0.62

 Pants

-0.61

 ASAP

-0.59

rous

-0.59

atche

-0.58

LESS

-0.57

oston

-0.56

POSITIVE LOGITS

 well

1.19

ylum

1.13

pired

1.12

 regards

1.12

pires

1.00

pects

0.97

phy

0.96

bestos

0.94

well

0.89

yl

0.89

Activations Density 0.111%

locations or settings

the word "as" in various contexts

No Comments

No Known Activations