INDEX

Explanations

terms related to structure and organization, particularly in data and categorization contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

imento

-0.16

COPE

-0.16

iol

-0.14

 Ø±ÙĪØ³ØªØ§

-0.14

 jose

-0.14

 Cave

-0.14

rone

-0.14

avÄĽ

-0.14

agged

-0.14

amen

-0.14

POSITIVE LOGITS

ipy

0.15

tk

0.15

zm

0.14

wig

0.14

 Goal

0.13

à¸¶à¸ģ

0.13

adata

0.13

 goal

0.13

.jsx

0.13

Activations Density 0.004%

terms related to structure and organization, particularly in data and categorization contexts

No Comments

No Known Activations