INDEX

Explanations

mentions of specific individuals and their roles or contributions, particularly in a work or organizational context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lost

-0.15

imbus

-0.15

iamond

-0.15

.getOwnProperty

-0.14

Unnamed

-0.14

vida

-0.14

Messenger

-0.14

eur

-0.14

ighbor

-0.14

.getFont

-0.14

POSITIVE LOGITS

 helped

0.29

 overs

0.27

 played

0.27

 successfully

0.22

ste

0.21

led

0.20

help

0.20

 broker

0.19

 Played

0.19

played

0.19

Activations Density 0.129%

mentions of specific individuals and their roles or contributions, particularly in a work or organizational context

No Comments

No Known Activations