INDEX

Explanations

questions about hypothetical scenarios or possibilities

oai_token-act-pair · gpt-3.5-turbo

speculative questions that begin with "What would."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

noticed

-0.67

haus

-0.66

held

-0.66

chio

-0.65

holm

-0.63

claimed

-0.63

illary

-0.60

trap

-0.60

cro

-0.60

 traced

-0.60

POSITIVE LOGITS

you

0.89

YOU

0.81

 anyone

0.77

 anybody

0.77

be

0.76

 motivate

0.75

 suffice

0.73

 happen

0.73

ĸļ

0.73

acan

0.72

Activations Density 0.041%

questions about hypothetical scenarios or possibilities

speculative questions that begin with "What would."

No Comments

No Known Activations