INDEX

Explanations

abbreviated names or initials of people and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

bach

-0.16

afort

-0.16

ANNOT

-0.15

_tC

-0.15

leon

-0.14

asmus

-0.14

ohana

-0.14

ohon

-0.14

onaut

-0.14

.gmail

-0.14

POSITIVE LOGITS

 corners

0.14

 Duck

0.14

tek

0.14

imensional

0.13

 Rose

0.13

urve

0.13

 (“

0.13

Ef

0.13

 Roller

0.13

Å¡tÄĽ

0.13

Activations Density 0.044%

abbreviated names or initials of people and organizations

No Comments

No Known Activations