INDEX

Explanations

words related to software development and improvement

oai_token-act-pair · gpt-3.5-turbo

terms related to improvements and modifications

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

Nazis

-1.02

urches

-1.00

lees

-0.95

Jews

-0.94

events

-0.92

Muslims

-0.91

apses

-0.90

deals

-0.88

storms

-0.87

pointers

-0.87

POSITIVE LOGITS

 version

1.90

 variant

1.44

 edition

1.39

 iteration

1.36

 copy

1.32

 incarnation

1.30

 counterpart

1.25

 replica

1.22

 example

1.22

 model

1.19

Activations Density 0.560%

words related to software development and improvement

terms related to improvements and modifications

No Comments

No Known Activations