INDEX

Explanations

time-related data points and numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

qc

-0.19

peare

-0.17

pac

-0.16

rances

-0.16

qc

-0.15

 Wheel

-0.15

uden

-0.15

ãģıãģł

-0.15

fas

-0.15

-0.14

POSITIVE LOGITS

ãĤ¤ãĥĪ

0.15

ego

0.14

.Interop

0.14

egot

0.14

roj

0.14

 agon

0.14

anny

0.14

luet

0.14

arp

0.14

ulong

0.14

Activations Density 0.163%

time-related data points and numerical values

No Comments

No Known Activations