INDEX

Explanations

comparisons between different options based on quantity or quality

oai_token-act-pair · gpt-3.5-turbo

quantitative terms and measurements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

CVE

-0.59

 vast

-0.53

acly

-0.53

 Jungle

-0.53

 slightest

-0.53

algia

-0.52

 Command

-0.52

arton

-0.52

 tiny

-0.52

 nutshell

-0.51

POSITIVE LOGITS

than

2.14

 than

1.99

 Than

1.40

 wiser

0.77

 Faster

0.74

 prevailed

0.72

 compared

0.71

models

0.71

iating

0.70

urations

0.69

Activations Density 0.329%

comparisons between different options based on quantity or quality

quantitative terms and measurements

No Comments

No Known Activations