INDEX

Explanations

mentions or references to echoing or repeating statements

oai_token-act-pair · gpt-3.5-turbo

references to previously mentioned ideas or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

abouts

-0.78

ONSORED

-0.72

bang

-0.72

 compensated

-0.70

dating

-0.69

 Leilan

-0.67

istan

-0.67

hess

-0.67

 Tanks

-0.63

abilia

-0.62

POSITIVE LOGITS

 tenets

1.13

 sentiments

1.08

 premise

1.08

 principles

1.02

 principle

1.02

 notion

1.00

 wishes

0.99

 norms

0.95

 recommendations

0.95

 teachings

0.94

Activations Density 0.510%

mentions or references to echoing or repeating statements

references to previously mentioned ideas or concepts

No Comments

No Known Activations