INDEX

Explanations

phrases related to security forces

oai_token-act-pair · gpt-3.5-turbo

terms related to security forces and services

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

ãĤ¨ãĥ«

-0.76

å§«

-0.76

undred

-0.74

Cola

-0.73

erent

-0.72

iasco

-0.71

zeb

-0.69

 Jelly

-0.68

 Cheese

-0.65

lio

-0.63

POSITIVE LOGITS

warts

0.83

illance

0.68

 Clapper

0.67

 fingerprints

0.66

iland

0.63

 Benghazi

0.62

rity

0.62

 infiltration

0.62

 powers

0.61

 Huawei

0.61

Activations Density 0.234%

phrases related to security forces

terms related to security forces and services

No Comments

No Known Activations