INDEX

Explanations

specific names or identifiers related to various fields and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

eger

-0.16

onse

-0.15

 Campos

-0.15

LF

-0.15

edic

-0.14

!=(

-0.14

inee

-0.14

ic

-0.14

icz

-0.13

onde

-0.13

POSITIVE LOGITS

Ã³s

0.15

oeff

0.14

ì´

0.14

onya

0.14

ihat

0.14

hausen

0.14

rops

0.14

iferay

0.14

reau

0.14

mass

0.13

Activations Density 0.019%

specific names or identifiers related to various fields and contexts

No Comments

No Known Activations