INDEX

Explanations

comparisons between replicas or simulations and their original counterparts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

baugh

-0.15

ãĤĵãģ¨

-0.13

 snad

-0.13

uel

-0.13

precated

-0.13

RYPT

-0.13

abis

-0.12

Å¯j

-0.12

Ø´ØªÙĩ

-0.12

 weiber

-0.12

POSITIVE LOGITS

 real

1.20

real

0.98

 REAL

0.91

 Real

0.88

Real

0.84

-real

0.83

_real

0.79

.real

0.79

(real

0.78

REAL

0.78

Activations Density 0.319%

comparisons between replicas or simulations and their original counterparts

No Comments

No Known Activations