INDEX

Explanations

references to quantity, measurement, and comparisons involving numbers or data

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

_overflow

-0.18

Map

-0.14

ili

-0.14

onia

-0.14

sim

-0.13

nton

-0.13

-uppercase

-0.13

orgia

-0.13

ij

-0.13

POSITIVE LOGITS

ÑĨÑİ

0.18

ICO

0.15

ahas

0.15

ico

0.15

Ret

0.14

 Baxter

0.14

zin

0.14

chte

0.14

uddle

0.14

yar

0.14

Activations Density 0.009%

references to quantity, measurement, and comparisons involving numbers or data

No Comments

No Known Activations