INDEX

Explanations

specific nouns and related terms associated with social issues and responsibilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Spi

-0.15

hana

-0.15

aned

-0.15

iedy

-0.14

laughter

-0.14

opers

-0.14

quier

-0.14

'ye

-0.14

ksam

-0.14

peech

-0.13

POSITIVE LOGITS

sor

0.17

ron

0.15

èĳ

0.15

 Intercept

0.14

 suppl

0.14

who

0.14

ek

0.14

pcs

0.14

igsaw

0.14

who

0.14

Activations Density 0.183%

specific nouns and related terms associated with social issues and responsibilities

No Comments

No Known Activations