INDEX

Explanations

words related to the mouth, such as actions involving the mouth and sensory experiences related to it

oai_token-act-pair · gpt-3.5-turbo

references to the word "mouth" 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 @ 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

Syd

-0.72

 Myst

-0.67

 Levi

-0.65

 Retro

-0.65

 Anthem

-0.63

 Samoa

-0.63

actionDate

-0.63

 Matrix

-0.63

 ACPI

-0.62

 Scotia

-0.61

POSITIVE LOGITS

piece

1.63

parts

1.49

pieces

1.46

wash

1.34

ful

1.22

feel

1.20

bone

1.09

pipe

1.07

bones

1.07

water

0.96

Activations Density 0.041%

words related to the mouth, such as actions involving the mouth and sensory experiences related to it

references to the word "mouth" in various contexts

No Comments

No Known Activations