INDEX

Explanations

phrases that introduce topics or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

otomy

-0.15

oret

-0.15

nd

-0.15

idel

-0.14

ket

-0.14

raith

-0.14

ards

-0.14

rena

-0.14

sg

-0.14

 vict

-0.14

POSITIVE LOGITS

aign

0.17

ãĥ¶

0.16

ayan

0.15

zby

0.14

gie

0.14

 REGARD

0.14

elda

0.14

obox

0.14

Nib

0.14

(EIF

0.13

Activations Density 0.015%

phrases that introduce topics or subjects

No Comments

No Known Activations