INDEX

Explanations

phrases indicating a decision or action step being taken

oai_token-act-pair · gpt-3.5-turbo

the word "so" used to indicate purpose or consequence

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

 glances

-0.62

ãģ£

-0.57

wreck

-0.57

 silhouette

-0.57

 Mens

-0.56

 disbelief

-0.56

 gallery

-0.56

 denial

-0.55

 realities

-0.55

 outline

-0.54

POSITIVE LOGITS

bered

1.24

othes

1.13

apy

1.10

oths

1.08

othe

1.07

aps

0.99

oner

0.98

oooo

0.90

arer

0.90

arin

0.87

Activations Density 0.092%

phrases indicating a decision or action step being taken

the word "so" used to indicate purpose or consequence

No Comments

No Known Activations