INDEX

Explanations

references to people's names and locations

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

 Anaheim

-0.17

orne

-0.16

 Patrol

-0.15

ç®¡

-0.15

 innocent

-0.15

fcn

-0.14

iami

-0.14

aghan

-0.14

 Midlands

-0.14

 Imperial

-0.14

POSITIVE LOGITS

 Charlottesville

0.19

 Organic

0.17

 organic

0.17

iasi

0.16

 Ober

0.16

emacs

0.16

coh

0.15

 Lyme

0.15

organic

0.15

(::

0.15

Activations Density 0.284%

references to people's names and locations

No Comments

No Known Activations