INDEX

Explanations

phrases related to violent actions

oai_token-act-pair · gpt-3.5-turbo

phrases related to violence and aggression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tycoon

-0.78

 Horus

-0.72

 charms

-0.72

 Europa

-0.70

 Scotia

-0.70

 modification

-0.69

 Confederation

-0.68

oller

-0.68

 Daniels

-0.68

 Vaughn

-0.67

POSITIVE LOGITS

paren

1.16

bodied

1.10

up

1.03

edge

1.02

out

1.01

by

1.00

average

0.99

upon

0.98

looking

0.97

together

0.97

Activations Density 0.051%

phrases related to violent actions

phrases related to violence and aggression

No Comments

No Known Activations