INDEX

Explanations

numerical identifiers or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

uÅ¡

-0.17

izr

-0.15

onu

-0.15

ãĢħ

-0.15

zyst

-0.14

 Î±ÏģÏĩ

-0.14

ldb

-0.14

ilden

-0.14

apons

-0.13

POSITIVE LOGITS

mAh

0.21

/QÄĲ

0.18

rpm

0.18

tics

0.16

MHz

0.15

-ci

0.15

Ø¡

0.15

adia

0.15

ses

0.15

à¤Ī

0.14

Activations Density 0.134%

numerical identifiers or codes

No Comments

No Known Activations