INDEX

Explanations

words related to qualitative descriptions, particularly concerning characteristics or attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

(sp

-0.16

âĢį

-0.15

“

-0.14

ÑĩÐ¸Ðº

-0.14

 Extra

-0.14

ingly

-0.14

ibly

-0.14

ment

-0.14

 Divine

-0.13

POSITIVE LOGITS

ãĢħ

0.16

ENCIL

0.16

.ht

0.16

-mÃªme

0.16

-Ãªtre

0.14

atitis

0.14

ylim

0.14

obuf

0.14

.gdx

0.14

ovol

0.14

Activations Density 0.260%

words related to qualitative descriptions, particularly concerning characteristics or attributes

No Comments

No Known Activations