INDEX

Explanations

common phrases or names related to morning talk shows

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Morning."

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

inates

-0.88

cial

-0.73

relative

-0.72

staking

-0.68

ificial

-0.68

mental

-0.68

Engineers

-0.67

kefeller

-0.67

icy

-0.67

POSITIVE LOGITS

 Caller

0.86

 Morning

0.86

night

0.84

star

0.83

abies

0.71

 Edition

0.70

odon

0.70

aturdays

0.68

undown

0.66

Dew

0.66

Activations Density 0.014%

common phrases or names related to morning talk shows

occurrences of the word "Morning."

No Comments

No Known Activations