INDEX

Explanations

references to robots

oai_token-act-pair · gpt-3.5-turbo

mentions of robots

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Lans

-0.78

etimes

-0.78

 Meadows

-0.77

Occ

-0.75

Collins

-0.72

 Mead

-0.71

 Byrd

-0.70

 Graves

-0.70

ptoms

-0.70

 Browne

-0.68

POSITIVE LOGITS

 robot

3.61

 robots

2.84

 Robot

2.49

 robotic

2.36

 Robots

2.25

 robotics

2.11

 droid

1.85

 bots

1.82

 Robotics

1.72

 Robo

1.68

Activations Density 0.025%

references to robots

mentions of robots

No Comments

No Known Activations