INDEX

Explanations

numerical values and specific data formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

éª¨

-0.17

illery

-0.16

ãĥ³ãĥĶ

-0.16

awi

-0.15

asco

-0.15

idel

-0.15

 Photon

-0.14

olta

-0.14

izzer

-0.14

clar

-0.14

POSITIVE LOGITS

å®¿

0.17

iy

0.16

0.15

ãĥ»ãĥ»ãĥ»↵↵

0.14

lete

0.14

AGMA

0.14

 Gree

0.14

CAPE

0.14

 Goldberg

0.14

Activations Density 0.001%

numerical values and specific data formats

No Comments

No Known Activations