INDEX

Explanations

references to database commands and their text parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 tackle

-0.16

 Tall

-0.14

 |--------------------------------------------------------------------------↵

-0.13

ToJson

-0.13

TransparentColor

-0.13

Tuple

-0.13

styleType

-0.13

Ð½ÐµÐ½

-0.13

soap

-0.12

Tile

-0.12

POSITIVE LOGITS

 text

1.29

 Text

1.08

text

1.05

 texts

0.99

Text

0.98

 TEXT

0.94

_text

0.93

.text

0.91

-text

0.91

 ÑĤÐµÐºÑģÑĤ

0.86

Activations Density 0.396%

references to database commands and their text parameters

No Comments

No Known Activations