INDEX

Explanations

mentions of military-related entities, leaders, and activities

oai_token-act-pair · gpt-3.5-turbo

names and organizations related to military and political topics

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

 THEN

-0.77

_______

-0.75

 attRot

-0.73

 Anyway

-0.71

ĺħ

-0.67

then

-0.67

rame

-0.66

wikipedia

-0.65

______

-0.64

brainer

-0.62

POSITIVE LOGITS

 continues

1.34

has

1.22

 remains

1.22

 hasn

1.16

 retains

1.16

 refuses

1.13

 maintains

1.12

 enjoys

1.11

 dominates

1.10

 spends

1.06

Activations Density 0.414%

mentions of military-related entities, leaders, and activities

names and organizations related to military and political topics

No Comments

No Known Activations