INDEX

Explanations

the concept of suffering, especially when related to exploitation or harm inflicted on individuals

oai_token-act-pair · gpt-3.5-turbo

references to the concept of suffering

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cle

-0.62

ococ

-0.62

 Shots

-0.61

sure

-0.61

RAFT

-0.60

ouncing

-0.60

ioch

-0.59

olid

-0.58

gat

-0.57

sports

-0.57

POSITIVE LOGITS

lehem

0.89

 inflicted

0.82

lessly

0.79

 Palest

0.76

hani

0.75

 suffering

0.72

 Nadu

0.71

nesses

0.70

 suffered

0.69

 setbacks

0.68

Activations Density 0.017%

the concept of suffering, especially when related to exploitation or harm inflicted on individuals

references to the concept of suffering

No Comments

No Known Activations