INDEX

Explanations

mentions of television network programming schedule or promotions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "on" and variations of the word "to"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Rebirth

-0.62

 blot

-0.62

 Paran

-0.61

 intervening

-0.60

 psychotic

-0.58

ĪĴ

-0.58

 mete

-0.57

 Crimes

-0.56

 Jericho

-0.56

 incent

-0.56

POSITIVE LOGITS

merce

0.87

opot

0.79

ĸļ

0.77

hing

0.74

umo

0.71

ouched

0.70

ulo

0.70

*=-

0.70

erman

0.70

ulum

0.70

Activations Density 0.063%

mentions of television network programming schedule or promotions

occurrences of the word "on" and variations of the word "to"

No Comments

No Known Activations