INDEX

Explanations

the word "permafrost" or variations of it

oai_token-act-pair · gpt-3.5-turbo

references to "permafrost" and related concepts

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

VERTIS

-0.81

yer

-0.80

roads

-0.75

lycer

-0.73

GER

-0.71

glers

-0.70

rider

-0.69

Bus

-0.68

llo

-0.67

GI

-0.67

POSITIVE LOGITS

utations

1.15

utation

1.01

ittee

0.99

 perm

0.98

iton

0.94

essage

0.94

itic

0.89

ilit

0.87

ittees

0.87

isphere

0.86

Activations Density 0.011%

the word "permafrost" or variations of it

references to "permafrost" and related concepts

No Comments

No Known Activations