INDEX

Explanations

terms related to focusing on specific details or examination at a close range

oai_token-act-pair · gpt-3.5-turbo

instances of the word "close" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

AIN

-0.84

schild

-0.67

 Pengu

-0.64

 Rosenstein

-0.64

ICAN

-0.63

onne

-0.62

 Kamp

-0.62

 Frie

-0.62

Sov

-0.61

STE

-0.61

POSITIVE LOGITS

 proximity

1.05

 examination

0.91

 shave

0.90

ups

0.90

 inspection

0.89

eness

0.89

paren

0.88

 scrutiny

0.87

 approximation

0.84

 resemblance

0.83

Activations Density 0.025%

terms related to focusing on specific details or examination at a close range

instances of the word "close" and its variations

No Comments

No Known Activations