INDEX

Explanations

occurrences of numerical representations or formatted strings

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

jb

-0.17

jk

-0.17

zb

-0.16

zas

-0.15

iano

-0.15

 Swinger

-0.14

zs

-0.14

jan

-0.14

Hav

-0.14

POSITIVE LOGITS

eki

0.15

commons

0.14

uddy

0.14

_procs

0.14

0.13

liÄį

0.13

Ä©

0.13

ãĥ³ãĥģ

0.13

AGR

0.13

ableView

0.13

Activations Density 0.007%

occurrences of numerical representations or formatted strings

No Comments

No Known Activations

occurrences of numerical representations or formatted strings

No Comments

No Known Activations