INDEX

Explanations

instances of the word 'text' or variations of it, such as 'texts' or 'texting'

oai_token-act-pair · gpt-3.5-turbo

mentions of texting and text messaging

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

Reviewer

-0.74

steen

-0.70

mite

-0.69

aldi

-0.67

luster

-0.67

vernment

-0.66

pure

-0.66

izont

-0.66

 Flavoring

-0.65

vre

-0.65

POSITIVE LOGITS

 messages

1.20

 messaging

1.09

 texting

1.01

 Messages

1.01

 message

0.94

message

0.93

ured

0.87

 texted

0.86

mess

0.83

 alerts

0.83

Activations Density 0.021%

instances of the word 'text' or variations of it, such as 'texts' or 'texting'

mentions of texting and text messaging

No Comments

No Known Activations