INDEX

Explanations

references to people, particularly in a historical or biographical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

izik

-0.18

-Key

-0.15

isks

-0.15

ubu

-0.14

:::

-0.14

ëĤ

-0.14

ius

-0.13

 Hook

-0.13

own

-0.13

lining

-0.13

POSITIVE LOGITS

ross

0.16

ADF

0.15

achers

0.15

ventions

0.15

TEL

0.15

 Dealer

0.15

acher

0.15

 Dealers

0.15

ê±¸

0.14

ast

0.14

Activations Density 0.026%

references to people, particularly in a historical or biographical context

No Comments

No Known Activations