INDEX

Explanations

words related to the concept of "night."

oai_token-act-pair · gpt-3.5-turbo

references to "Night" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xon

-0.80

achev

-0.76

pta

-0.75

 sqor

-0.71

igion

-0.70

hetical

-0.70

ä½ľ

-0.69

ickr

-0.69

kson

-0.68

POSITIVE LOGITS

mares

1.04

club

1.01

bird

0.89

mare

0.88

fall

0.83

 Night

0.82

night

0.81

 Riders

0.81

Owl

0.81

Shy

0.79

Activations Density 0.017%

words related to the concept of "night."

references to "Night" in various contexts

No Comments

No Known Activations