INDEX

Explanations

words related to referring or naming something

oai_token-act-pair · gpt-3.5-turbo

phrases that contain the word "referred" indicating descriptions or references to other entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ï¸ı

-0.89

alde

-0.79

iem

-0.66

iasm

-0.66

mo

-0.66

nesia

-0.65

 stoked

-0.64

raged

-0.63

ointment

-0.63

 pitted

-0.63

POSITIVE LOGITS

 hereafter

0.75

 initials

0.74

ãĥ¼ãĥ³

0.72

him

0.68

 them

0.68

 herein

0.67

Him

0.66

"-

0.66

Hud

0.66

 terminology

0.65

Activations Density 0.083%

words related to referring or naming something

phrases that contain the word "referred" indicating descriptions or references to other entities

No Comments

No Known Activations