INDEX

Explanations

phrases related to military operations and organizations

oai_token-act-pair · gpt-3.5-turbo

references to awareness and specific groups or entities within conflict situations

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

Russ

-0.73

 Columbus

-0.68

Nic

-0.66

 favors

-0.63

Â¯Â¯

-0.60

 Sonic

-0.60

 admitting

-0.60

DUI

-0.59

lean

-0.59

Sil

-0.58

POSITIVE LOGITS

awar

5.07

merga

1.67

asar

1.51

arak

1.16

ahar

1.12

urat

1.07

 Karachi

1.05

airo

1.04

 Kham

1.03

bah

1.01

Activations Density 0.030%

phrases related to military operations and organizations

references to awareness and specific groups or entities within conflict situations

No Comments

No Known Activations