INDEX

Explanations

references to differential concepts and their variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

warm

-0.15

\Collections

-0.15

 háºŃu

-0.14

inee

-0.14

inyin

-0.14

achie

-0.14

_utf

-0.14

pliant

-0.14

abic

-0.14

ouble

-0.14

POSITIVE LOGITS

0.20

rent

0.20

ulty

0.18

rence

0.18

 diagnosis

0.17

 calculus

0.17

renc

0.17

iator

0.15

Eg

0.15

bedo

0.15

Activations Density 0.012%

references to differential concepts and their variations in different contexts

No Comments

No Known Activations