INDEX

Explanations

HTML tags used for formatting and clickable links

oai_token-act-pair · gpt-3.5-turbo

HTML tags and their attributes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 satell

-0.78

cius

-0.77

 undermin

-0.75

ĪĴ

-0.74

etheless

-0.74

 isot

-0.72

 clin

-0.71

 unsolved

-0.68

¥ŀ

-0.67

 moder

-0.66

POSITIVE LOGITS

âĸĳ

0.92

{{

0.83

</

0.80

tab

0.78

 PsyNet

0.77

();

0.77

TPS

0.77

({

0.73

 Actions

0.73

"},{"

0.69

Activations Density 0.009%

HTML tags used for formatting and clickable links

HTML tags and their attributes

No Comments

No Known Activations