INDEX

Explanations

the word "lost" or mentions involving being lost

oai_token-act-pair · gpt-3.5-turbo

instances of the word "lost" or variations thereof

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

sis

-0.77

ERAL

-0.71

erity

-0.66

ENG

-0.66

NVIDIA

-0.65

instein

-0.65

UP

-0.64

cised

-0.63

Break

-0.63

CU

-0.62

POSITIVE LOGITS

 souls

1.02

 Souls

0.91

 sight

0.87

 innocence

0.86

luster

0.84

hearted

0.82

iscover

0.81

lust

0.80

 Tycoon

0.80

 virginity

0.79

Activations Density 0.036%

the word "lost" or mentions involving being lost

instances of the word "lost" or variations thereof

No Comments

No Known Activations