INDEX

Explanations

instances of the word "top"

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ob

-0.15

rl

-0.15

.ll

-0.15

/host

-0.15

bag

-0.14

zM

-0.14

 midfield

-0.14

ulla

-0.14

Rol

-0.14

ign

-0.13

POSITIVE LOGITS

lashes

0.17

Æ°á»Ľc

0.16

allis

0.15

onto

0.14

 latter

0.14

hel

0.14

ikler

0.14

cedes

0.14

kaar

0.14

kip

0.14

Activations Density 0.028%

instances of the word "top"

No Comments

No Known Activations