INDEX

Explanations

numerical references to volumes and issues of a publication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

[at

-0.16

ĵį

-0.15

ippi

-0.14

ived

-0.14

 Seconds

-0.14

ives

-0.14

ikon

-0.14

ource

-0.13

esto

-0.13

ÑĢÐ¾Ðº

-0.13

POSITIVE LOGITS

Iss

0.35

Iss

0.28

 issu

0.28

iss

0.26

iss

0.23

ISS

0.22

 issuing

0.21

Issuer

0.21

 issued

0.21

 issuer

0.21

Activations Density 0.008%

numerical references to volumes and issues of a publication

No Comments

No Known Activations