INDEX

Explanations

references to additional or alternative items and concepts

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

PFN

-0.17

udging

-0.17

ero

-0.16

kus

-0.16

osy

-0.15

eros

-0.15

itsu

-0.15

auce

-0.14

eren

-0.14

eres

-0.14

POSITIVE LOGITS

 dear

0.15

 others

0.15

IGHL

0.15

 Bench

0.14

UT

0.14

,module

0.14

 both

0.13

lez

0.13

 combinations

0.13

both

0.13

Activations Density 0.037%

references to additional or alternative items and concepts

No Comments

No Known Activations