INDEX

Explanations

comparative expressions related to quantity or degree

oai_token-act-pair · gpt-3.5-turbo

comparative terms indicating a decrease or lesser amount

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Origin

-0.85

Draft

-0.78

ãĤ±

-0.75

 Origins

-0.68

DK

-0.68

MA

-0.67

Trivia

-0.67

 Classification

-0.66

ORY

-0.66

Phase

-0.66

POSITIVE LOGITS

 than

1.13

 forgiving

0.89

 stringent

0.82

 manageable

0.81

 costly

0.80

 viable

0.77

 conspicuous

0.76

 profitable

0.76

ened

0.76

 sparing

0.75

Activations Density 0.025%

comparative expressions related to quantity or degree

comparative terms indicating a decrease or lesser amount

No Comments

No Known Activations