INDEX

Explanations

verbs that indicate actions or preferences related to enjoyment and participation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

allow

-0.16

ina

-0.15

 allow

-0.15

feit

-0.14

 today

-0.14

jt

-0.14

 insensitive

-0.14

ismet

-0.14

ift

-0.13

 possibly

-0.13

POSITIVE LOGITS

SetActive

0.15

 ìĿĳ

0.15

IMER

0.15

 Druh

0.15

iyi

0.14

nez

0.14

sie

0.14

Ð½Ð°ÑĢ

0.14

SError

0.14

Activations Density 0.038%

verbs that indicate actions or preferences related to enjoyment and participation

No Comments

No Known Activations