INDEX

Explanations

phrases that present conditions or hypotheticals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ili

-0.20

oster

-0.18

_bn

-0.15

CRET

-0.15

enk

-0.15

lox

-0.14

 undert

-0.14

esco

-0.14

tridge

-0.13

ÑģÑĤÐ¸

-0.13

POSITIVE LOGITS

ebb

0.15

 Shea

0.14

SPAN

0.14

 Ð¿Ð¾ÑĪ

0.14

 Tanner

0.14

AUDIO

0.13

CAC

0.13

Vu

0.13

osci

0.13

PEC

0.13

Activations Density 0.021%

phrases that present conditions or hypotheticals

No Comments

No Known Activations