INDEX

Explanations

phrases that instruct or encourage readers to engage with content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

oret

-0.17

acades

-0.15

ono

-0.14

quet

-0.14

ieved

-0.14

ì¼ĵ

-0.14

 McGr

-0.14

STRING

-0.13

ãĤ¯ãĥĪ

-0.13

iska

-0.13

POSITIVE LOGITS

erdale

0.18

 Gratis

0.16

ÑĢÐ°ÑĤÐ¸

0.15

erus

0.15

otte

0.14

Ð»ÑĸÐ²

0.14

ãĥĥãĥĦ

0.14

Ð¾ÑĢÐ´

0.14

ITAL

0.14

ADDE

0.14

Activations Density 0.027%

phrases that instruct or encourage readers to engage with content

No Comments

No Known Activations