INDEX

Explanations

specific names, possibly of a person named "Robinson."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name 'Robinson'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uate

-0.86

uated

-0.81

SEE

-0.81

ablishment

-0.80

ysis

-0.79

WAYS

-0.77

rolog

-0.71

ropolitan

-0.70

uates

-0.69

umbers

-0.69

POSITIVE LOGITS

inson

1.18

 Crus

1.17

 Robinson

0.86

 Mansion

0.70

bie

0.70

otics

0.67

 Armory

0.65

 Hardware

0.63

ç·

0.61

 Saunders

0.61

Activations Density 0.019%

specific names, possibly of a person named "Robinson."

occurrences of the name 'Robinson'

No Comments

No Known Activations