INDEX

Explanations

specific structured elements or categories in written content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

pie

-0.16

bob

-0.15

ANNEL

-0.15

untime

-0.14

.getExternal

-0.14

 Sims

-0.14

 Sachs

-0.14

.dylib

-0.14

ulta

-0.13

_CAMERA

-0.13

POSITIVE LOGITS

uman

0.16

/on

0.15

elage

0.15

VERSE

0.14

uyo

0.14

/by

0.14

Laf

0.14

 ï½ľ

0.14

 Phelps

0.14

uye

0.13

Activations Density 0.065%

specific structured elements or categories in written content

No Comments

No Known Activations