INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to the word "lad" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ELL

-0.83

kward

-0.69

plays

-0.67

forts

-0.67

EED

-0.67

basketball

-0.66

 Bakr

-0.64

 ticking

-0.64

ä¸Ń

-0.63

oldown

-0.62

POSITIVE LOGITS

imir

1.41

der

1.04

isl

0.93

mir

0.92

ynam

0.92

itionally

0.85

itional

0.82

amus

0.80

ewater

0.79

rian

0.78

Activations Density 0.036%

names of individuals

references to the word "lad" and related variations

No Comments

No Known Activations