INDEX

Explanations

concepts related to instruction and guidance

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

err

-0.16

ifest

-0.14

ernet

-0.14

ofilm

-0.13

oky

-0.13

loff

-0.13

znik

-0.13

arc

-0.13

PIN

-0.13

ACY

-0.13

POSITIVE LOGITS

gunta

0.14

.wh

0.14

ëł¨

0.14

Ymd

0.13

.intellij

0.13

atÃ³

0.13

 Ð¸Ð¼Ñĥ

0.13

 Towers

0.13

Broad

0.13

Activations Density 0.101%

concepts related to instruction and guidance

No Comments

No Known Activations