INDEX

Explanations

time-related phrases, specifically mentioning weeks or months

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate time intervals or durations

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

UTF

-0.70

ãĤ¨ãĥ«

-0.68

 Mechdragon

-0.67

Sph

-0.66

SourceFile

-0.64

 Scully

-0.64

 Sovere

-0.63

 counterfe

-0.63

 Fram

-0.61

utf

-0.61

POSITIVE LOGITS

 devoted

0.83

 probation

0.79

ONY

0.79

iversary

0.76

 apiece

0.73

isode

0.73

idays

0.72

 elapsed

0.71

trip

0.69

onies

0.69

Activations Density 0.245%

time-related phrases, specifically mentioning weeks or months

phrases that indicate time intervals or durations

No Comments

No Known Activations