INDEX

Explanations

concepts related to accountability and obligation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ATING

-0.17

ATED

-0.15

wik

-0.14

ated

-0.14

acional

-0.14

avec

-0.14

ATORS

-0.14

Åĳs

-0.14

SPATH

-0.14

Î³Î¬

-0.14

POSITIVE LOGITS

ment

0.36

ance

0.33

rence

0.28

ishment

0.26

ption

0.26

ization

0.25

ence

0.24

gence

0.24

ure

0.23

iture

0.23

Activations Density 0.067%

concepts related to accountability and obligation

No Comments

No Known Activations