INDEX

Explanations

mentions or references to the name "Duncan"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Duncan."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Magikarp

-0.81

-0.80

ocard

-0.72

 hypers

-0.68

xual

-0.67

ocene

-0.66

othal

-0.65

juven

-0.65

ī

-0.63

 Mandal

-0.63

POSITIVE LOGITS

 Duncan

1.17

querque

0.84

Leaks

0.84

inet

0.83

geon

0.83

ridge

0.82

 Smith

0.77

aii

0.76

ienne

0.75

Smy

0.75

Activations Density 0.005%

mentions or references to the name "Duncan"

mentions of the name "Duncan."

No Comments

No Known Activations