INDEX

Explanations

actions related to violence

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate actions or intentions related to "to" phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Observer

-0.63

 Bhar

-0.60

 Parameters

-0.60

 employer

-0.59

 Trek

-0.59

 traveller

-0.59

IR

-0.59

 Country

-0.57

 traveler

-0.55

 franch

-0.54

POSITIVE LOGITS

 commemorate

1.28

pless

1.13

 simulate

1.12

 alleviate

1.11

 celebrate

1.09

 relieve

1.07

 signify

1.06

wered

1.04

asted

1.04

 emphasize

1.03

Activations Density 0.230%

actions related to violence

phrases that indicate actions or intentions related to "to" phrases

No Comments

No Known Activations