INDEX

Explanations

language related to strengthening, fortifying, or supporting something

oai_token-act-pair · gpt-3.5-turbo

terms related to reinforcement and its effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

chens

-0.81

«ĺ

-0.77

kers

-0.74

soever

-0.72

izons

-0.71

kie

-0.69

cture

-0.68

NetMessage

-0.66

anas

-0.64

hair

-0.64

POSITIVE LOGITS

 reinforcing

1.36

 reinforcement

1.10

 reinforced

0.98

 reinforce

0.88

 additive

0.81

 reinforces

0.80

forcement

0.79

irmation

0.76

ments

0.72

forcing

0.70

Activations Density 0.014%

language related to strengthening, fortifying, or supporting something

terms related to reinforcement and its effects

No Comments

No Known Activations