INDEX

Explanations

words related to mental health or psychological states, specifically with a focus on negative connotations

oai_token-act-pair · gpt-3.5-turbo

terms associated with irrational or extreme behavior, particularly in a political context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

âĢ¢âĢ¢

-0.75

 Norn

-0.75

ORN

-0.74

Constructed

-0.72

LECT

-0.72

USE

-0.71

åĬ

-0.70

 Dome

-0.69

EVA

-0.67

 Gamble

-0.66

POSITIVE LOGITS

cheon

1.22

acy

1.19

acies

1.17

atic

1.14

atics

1.10

atical

1.05

gement

1.03

lun

1.01

aries

0.98

atural

0.94

Activations Density 0.014%

words related to mental health or psychological states, specifically with a focus on negative connotations

terms associated with irrational or extreme behavior, particularly in a political context

No Comments

No Known Activations