INDEX

Explanations

specific types of labels or tags associated with various subjects, particularly those related to customer profiles, forms, and extensions in a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

vu

-0.17

ernet

-0.16

aler

-0.15

 Stout

-0.15

inst

-0.15

alam

-0.15

Æ¯

-0.14

enes

-0.13

 Presidential

-0.13

 repeat

-0.13

POSITIVE LOGITS

ÙĪÙĦÙĩ

0.15

ë¨¸ëĭĪ

0.15

omba

0.15

ONA

0.15

 Oswald

0.14

æ¯Ľ

0.14

lige

0.13

tram

0.13

 bers

0.13

LEMENT

0.13

Activations Density 0.067%

specific types of labels or tags associated with various subjects, particularly those related to customer profiles, forms, and extensions in a technical context

No Comments

No Known Activations