INDEX

Explanations

references to the action of rolling something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Roll" and its variations

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

ETH

-0.69

 diver

-0.67

 paras

-0.66

 parasitic

-0.66

 parasite

-0.66

assian

-0.65

sov

-0.65

 unbearable

-0.65

 resp

-0.64

 comprehension

-0.64

POSITIVE LOGITS

 Roll

4.04

Roll

2.78

 roll

2.37

roll

2.20

 Rolls

2.13

 Roller

1.95

 rolls

1.84

 rolled

1.61

 Rolling

1.58

 rolling

1.58

Activations Density 0.017%

references to the action of rolling something

occurrences of the word "Roll" and its variations

No Comments

No Known Activations