INDEX

Explanations

terms and expressions related to comparison and evaluation

oai_token-act-pair · gpt-3.5-turbo

references to "terms" related to comparisons or evaluations

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

resent

-0.84

avorite

-0.81

estern

-0.76

oaded

-0.76

 tatt

-0.71

oster

-0.70

 Towns

-0.70

illard

-0.70

enegger

-0.67

redited

-0.65

POSITIVE LOGITS

cale

0.89

pace

0.87

models

0.79

manship

0.79

cape

0.78

pring

0.76

terms

0.76

 terms

0.76

flation

0.75

hips

0.75

Activations Density 0.016%

terms and expressions related to comparison and evaluation

references to "terms" related to comparisons or evaluations

No Comments

No Known Activations