INDEX

Explanations

mentions of specific individuals, their actions, and personal details

oai_token-act-pair · gpt-3.5-turbo

references to significant events and their 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

apo

-0.92

Skin

-0.87

Baltimore

-0.85

oultry

-0.84

raviolet

-0.83

resist

-0.81

UV

-0.81

Gy

-0.80

Chicken

-0.79

Davis

-0.78

POSITIVE LOGITS

Cob

1.52

Dot

1.19

 Kurt

1.13

 Kand

1.13

 Titan

1.13

 Bent

1.11

 Hend

1.08

 Cout

1.06

 Cory

1.02

Dew

1.02

Activations Density 0.446%

mentions of specific individuals, their actions, and personal details

references to significant events and their implications

No Comments

No Known Activations