INDEX

Explanations

questions starting with "Would" and presenting hypothetical scenarios

oai_token-act-pair · gpt-3.5-turbo

questions posed to the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤĬ

-0.67

natureconservancy

-0.61

SPONSORED

-0.60

ãģĮ

-0.59

minus

-0.58

displayText

-0.58

ãĢĤ

-0.58

ãģ«

-0.57

hig

-0.57

 traced

-0.57

POSITIVE LOGITS

 anyone

1.04

1.03

 anybody

1.02

you

0.94

it

0.86

 they

0.83

 somebody

0.82

we

0.81

 someone

0.81

YOU

0.74

Activations Density 0.074%

questions starting with "Would" and presenting hypothetical scenarios

questions posed to the reader

No Comments

No Known Activations