INDEX

Explanations

references to specific laboratories and research facilities

oai_token-act-pair · gpt-3.5-turbo

references to various laboratories and research facilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

theless

-0.81

cuts

-0.71

BOOK

-0.68

making

-0.67

Args

-0.66

conv

-0.65

arching

-0.63

board

-0.60

 religions

-0.60

\":

-0.59

POSITIVE LOGITS

 Laboratories

1.06

 Labs

1.04

orer

1.03

rador

0.96

 Associates

0.91

rats

0.83

cius

0.81

ateur

0.78

 labs

0.77

ource

0.76

Activations Density 0.011%

references to specific laboratories and research facilities

references to various laboratories and research facilities

No Comments

No Known Activations