INDEX

Explanations

references to entries in catalogs or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

altar

-0.16

aley

-0.15

æĢģ

-0.15

arcer

-0.14

dart

-0.14

amura

-0.14

ward

-0.14

_INITIALIZER

-0.14

sterdam

-0.14

ë¡ľëĵľ

-0.13

POSITIVE LOGITS

ues

0.74

uing

0.60

ued

0.58

ue

0.54

uer

0.48

uers

0.46

UES

0.44

uem

0.43

ueue

0.41

UE

0.41

Activations Density 0.019%

references to entries in catalogs or lists

No Comments

No Known Activations