INDEX

Explanations

words related to physical or abstract stretches

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stretch" in various contexts

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

imer

-0.68

inst

-0.67

 Cance

-0.66

lia

-0.65

ãĥ¯

-0.65

En

-0.64

Message

-0.64

 Emer

-0.61

 disappro

-0.61

informed

-0.60

POSITIVE LOGITS

 stretch

4.15

 Stretch

2.73

Stretch

2.18

 stretches

2.13

retch

1.85

 stretching

1.72

 stretched

1.66

 bend

1.32

 swell

1.16

 span

1.13

Activations Density 0.005%

words related to physical or abstract stretches

occurrences of the word "stretch" in various contexts

No Comments

No Known Activations