INDEX

Explanations

phrases related to narration within a story

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences or opinions

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

 Amen

-0.71

advertising

-0.64

-+

-0.62

 warr

-0.61

 Pengu

-0.59

 tamp

-0.59

ularity

-0.59

uum

-0.58

 sshd

-0.57

 Short

-0.56

POSITIVE LOGITS

ldon

0.70

hung

0.67

çīĪ

0.67

asion

0.65

GROUND

0.63

LOAD

0.63

atic

0.63

forth

0.62

abad

0.61

ICS

0.61

Activations Density 0.155%

phrases related to narration within a story

references to personal experiences or opinions

No Comments

No Known Activations