INDEX

Explanations

institutions or regions, specifically focusing on geopolitical and encompassing details

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in" within various contexts

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

%%

-0.74

CLASSIFIED

-0.74

Null

-0.65

killed

-0.64

GROUND

-0.63

#$

-0.62

 PLEASE

-0.62

Voice

-0.61

username

-0.61

awa

-0.60

POSITIVE LOGITS

 regards

1.18

 lieu

1.16

 accordance

1.16

 favor

1.14

 terms

1.12

 favour

1.10

 spite

1.10

clus

1.08

 order

1.08

 vitro

1.07

Activations Density 0.549%

institutions or regions, specifically focusing on geopolitical and encompassing details

occurrences of the word "in" within various contexts

No Comments

No Known Activations