INDEX

Explanations

descriptions of actions related to war or conflict

oai_token-act-pair · gpt-3.5-turbo

references to conflicts and the consequences faced by affected communities

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

 congr

-1.03

itialized

-0.84

 ðŁĳ

-0.79

STEM

-0.77

Thumbnail

-0.76

_______

-0.73

(/

-0.73

âĺħâĺħ

-0.73

(*

-0.72

(&

-0.72

POSITIVE LOGITS

 sometimes

1.16

sometimes

1.15

 often

1.11

 smugglers

1.10

often

1.09

 impunity

1.07

 smuggled

1.05

 informants

1.04

 routinely

1.03

 occasionally

1.01

Activations Density 0.623%

descriptions of actions related to war or conflict

references to conflicts and the consequences faced by affected communities

No Comments

No Known Activations