INDEX

Explanations

time-related concepts or things that occurred repeatedly

oai_token-act-pair · gpt-3.5-turbo

references to the number of times something has occurred or been achieved

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

amera

-0.88

haar

-0.82

sterdam

-0.76

ennes

-0.76

het

-0.71

anqu

-0.70

ourt

-0.69

imilar

-0.69

riad

-0.69

abase

-0.68

POSITIVE LOGITS

 Bucc

0.75

etheless

0.67

 veteran

0.65

CFL

0.64

ago

0.63

 underdog

0.63

 anniversary

0.63

 heavyweight

0.62

 traveller

0.62

 recording

0.61

Activations Density 0.033%

time-related concepts or things that occurred repeatedly

references to the number of times something has occurred or been achieved

No Comments

No Known Activations