INDEX

Explanations

words that indicate presence, participation, or involvement in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

icken

-0.16

itchen

-0.15

resher

-0.15

uisse

-0.14

otherwise

-0.14

icher

-0.14

yles

-0.14

ialect

-0.14

Å¡ÃŃ

-0.14

oton

-0.13

POSITIVE LOGITS

 industry

0.20

Industry

0.19

 Industry

0.19

stal

0.16

industry

0.16

 indeed

0.15

 healthcare

0.15

 wider

0.14

mol

0.14

Ã³l

0.14

Activations Density 0.006%

words that indicate presence, participation, or involvement in various contexts

No Comments

No Known Activations