INDEX

Explanations

words related to the concept of "down"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Down."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ami

-0.68

oka

-0.66

obi

-0.63

ollower

-0.61

ici

-0.61

endi

-0.59

atsuki

-0.58

 apparatus

-0.58

 exempt

-0.58

ensibly

-0.57

POSITIVE LOGITS

 Down

3.85

Down

2.61

 Downs

2.50

down

1.94

downs

1.63

 DOWN

1.63

 down

1.59

 downs

1.55

Up

1.45

DOWN

1.38

Activations Density 0.007%

words related to the concept of "down"

occurrences of the word "Down."

No Comments

No Known Activations