INDEX

Explanations

references to individuals with the initial 'R' and potentially their associated titles or positions

oai_token-act-pair · gpt-3.5-turbo

the character "R" 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

 Surprise

-0.70

Cue

-0.69

 ours

-0.66

 Schumer

-0.66

 diamonds

-0.65

 wills

-0.65

 concessions

-0.65

 fulfillment

-0.63

 worrying

-0.60

 stride

-0.60

POSITIVE LOGITS

oddy

1.39

ishi

1.28

hett

1.27

afi

1.25

ansom

1.24

ohan

1.20

agh

1.19

olf

1.18

itchie

1.18

alf

1.18

Activations Density 0.028%

references to individuals with the initial 'R' and potentially their associated titles or positions

the character "R" in various contexts

No Comments

No Known Activations