INDEX

Explanations

descriptive language related to interpersonal interactions and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 organis

-0.20

 organisation

-0.18

 unmist

-0.18

 neighbourhood

-0.17

 recognised

-0.17

 scept

-0.17

bc

-0.16

 ÙģÙī

-0.16

 Honour

-0.16

 organisers

-0.16

POSITIVE LOGITS

 apart

0.22

Sorted

0.20

 Apart

0.19

Apart

0.19

 nearly

0.17

 mate

0.17

sorted

0.17

 cheers

0.17

[color

0.16

nev

0.16

Activations Density 0.143%

descriptive language related to interpersonal interactions and recommendations

No Comments

No Known Activations