INDEX

Explanations

specific times of day mentioned in text

oai_token-act-pair · gpt-3.5-turbo

specific times related to events

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

 achievements

-0.68

 biography

-0.65

lov

-0.62

phr

-0.62

 NEVER

-0.62

 genius

-0.61

FAQ

-0.61

ollah

-0.61

ospons

-0.60

 subsc

-0.60

POSITIVE LOGITS

morning

0.81

cture

0.78

opers

0.77

ixties

0.76

usting

0.74

 gunshots

0.73

 motorcycles

0.72

âĸĪâĸĪ

0.70

 curfew

0.70

 sidx

0.69

Activations Density 0.195%

specific times of day mentioned in text

specific times related to events

No Comments

No Known Activations