INDEX

Explanations

words related to precision and correctness

oai_token-act-pair · gpt-3.5-turbo

words related to appropriation and its implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Job

-0.71

 Panic

-0.71

Aad

-0.69

 abundantly

-0.68

hops

-0.66

 Imran

-0.66

BuyableInstoreAndOnline

-0.65

Ħ¢

-0.63

 Noir

-0.63

 Jobs

-0.61

POSITIVE LOGITS

eness

1.28

ivism

1.27

iously

1.20

ivo

1.19

ivist

1.19

rices

1.11

iven

1.10

ious

1.08

IVES

1.05

ivities

1.02

Activations Density 0.073%

words related to precision and correctness

words related to appropriation and its implications

No Comments

No Known Activations