INDEX

Explanations

text referring to written content, such as articles or posts

oai_token-act-pair · gpt-3.5-turbo

references to articles or publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãħĭ

-0.67

ibly

-0.66

cffffcc

-0.66

ãĤ¦ãĤ¹

-0.65

aura

-0.65

ãħĭãħĭ

-0.64

miss

-0.61

iously

-0.61

antly

-0.60

itive

-0.60

POSITIVE LOGITS

 however

0.98

we

0.90

 which

0.88

 titled

0.85

 meanwhile

0.84

 moreover

0.80

available

0.75

 authors

0.74

 researchers

0.74

 there

0.73

Activations Density 0.203%

text referring to written content, such as articles or posts

references to articles or publications

No Comments

No Known Activations