INDEX

Explanations

action verbs that indicate ability or involvement in specific activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

strap

-0.15

lil

-0.14

urve

-0.14

paring

-0.13

ãĥ¬ãĤ¹

-0.13

rend

-0.13

 Kraj

-0.13

edl

-0.13

.AD

-0.13

erto

-0.13

POSITIVE LOGITS

row

0.15

arro

0.15

retch

0.14

iddy

0.13

 latina

0.13

.matcher

0.13

 losing

0.13

 middle

0.13

 west

0.13

 abre

0.13

Activations Density 0.289%

action verbs that indicate ability or involvement in specific activities

No Comments

No Known Activations