INDEX

Explanations

terms related to being late or lateness

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "late" in various contexts

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

edIn

-0.84

alach

-0.83

roma

-0.83

ikk

-0.78

utenberg

-0.77

ebook

-0.75

\\\\\\\\

-0.75

adr

-0.74

qua

-0.73

orius

-0.71

POSITIVE LOGITS

 afternoon

1.01

 nights

0.84

 enough

0.80

 evening

0.80

 evenings

0.79

 night

0.79

 antiquity

0.72

 morning

0.71

 innings

0.70

 hours

0.69

Activations Density 0.016%

terms related to being late or lateness

occurrences of the word "late" in various contexts

No Comments

No Known Activations