INDEX

Explanations

words related to creating or producing something

oai_token-act-pair · gpt-3.5-turbo

terms related to media and storytelling

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Lep

-0.93

Lev

-0.87

-0.84

 Levine

-0.83

Hill

-0.83

-0.80

Harris

-0.80

Lev

-0.80

zsche

-0.80

POSITIVE LOGITS

make

1.06

 Made

1.00

 MAKE

0.99

 maker

0.97

 Make

0.95

 make

0.94

made

0.93

 Maker

0.91

 Making

0.90

making

0.87

Activations Density 0.273%

words related to creating or producing something

terms related to media and storytelling

No Comments

No Known Activations