INDEX

Explanations

mentions of locations or living situations

oai_token-act-pair · gpt-3.5-turbo

the preposition "in."

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

casters

-0.73

ector

-0.69

Listener

-0.65

 incumb

-0.65

 ende

-0.65

 guiActiveUn

-0.64

ERROR

-0.63

¶æ

-0.63

killed

-0.63

llor

-0.63

POSITIVE LOGITS

 accordance

1.16

ordinate

1.13

animate

1.08

 lieu

1.07

 conjunction

1.00

 front

1.00

bound

1.00

 spite

0.97

 vitro

0.97

organic

0.97

Activations Density 0.384%

mentions of locations or living situations

the preposition "in."

No Comments

No Known Activations