INDEX

Explanations

places and locations, particularly those involved in conflict and military operations

oai_token-act-pair · gpt-3.5-turbo

names of places and locations involved in conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

TPP

-0.93

Netflix

-0.90

CAN

-0.76

laughter

-0.76

strom

-0.76

BILITIES

-0.76

NZ

-0.75

Native

-0.75

SON

-0.75

wcs

-0.74

POSITIVE LOGITS

 Homs

1.46

 Aleppo

1.45

 Idlib

1.44

 Raqqa

1.41

 Mosul

1.41

 Damascus

1.34

 Sheikh

1.30

 Tripoli

1.25

Kaf

1.23

 Beirut

1.21

Activations Density 0.203%

places and locations, particularly those involved in conflict and military operations

names of places and locations involved in conflicts

No Comments

No Known Activations