INDEX

Explanations

phrases related to significant events or operations

oai_token-act-pair · gpt-3.5-turbo

the word "Deep" and its related derivatives 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 tremend

-0.82

lessly

-0.81

 inev

-0.78

 awoken

-0.76

voc

-0.75

 crawl

-0.73

 scaff

-0.72

 ultr

-0.71

 clock

-0.71

quo

-0.70

POSITIVE LOGITS

 Foods

1.11

 Moves

1.10

Mind

1.08

 Feet

1.07

 Places

1.05

 Mind

1.01

 Ones

1.00

 Hands

0.99

Choice

0.98

 Coat

0.98

Activations Density 0.165%

phrases related to significant events or operations

the word "Deep" and its related derivatives in various contexts

No Comments

No Known Activations