INDEX

Explanations

terms related to valuable, historically significant objects or creations, such as artifacts and relics

oai_token-act-pair · gpt-3.5-turbo

mentions of artifacts and relics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

sen

-0.75

NN

-0.73

creen

-0.68

hua

-0.66

secut

-0.66

 coli

-0.65

rol

-0.63

ppers

-0.63

ationally

-0.62

Western

-0.62

POSITIVE LOGITS

 artifact

1.47

 artifacts

1.44

 arte

1.34

ifacts

1.33

artifacts

1.33

 Artifact

0.96

 relics

0.95

ifact

0.94

smanship

0.91

ibles

0.85

Activations Density 0.008%

terms related to valuable, historically significant objects or creations, such as artifacts and relics

mentions of artifacts and relics

No Comments

No Known Activations