INDEX

Explanations

instances of the word "tell" and its variations, indicating a focus on communication or directives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

olicit

-0.16

 ÑģÐ¿ÑĢÐ°Ð²

-0.16

åįļçī©

-0.16

ç½

-0.15

ifest

-0.15

-reference

-0.14

 trá»Ŀi

-0.14

ladu

-0.14

anova

-0.14

addir

-0.14

POSITIVE LOGITS

ëĶ°

0.16

us

0.16

okie

0.16

atcher

0.14

ur

0.14

ãĥĪãĥ«

0.14

abyrinth

0.14

IGENCE

0.14

 reap

0.13

Activations Density 0.061%

instances of the word "tell" and its variations, indicating a focus on communication or directives

No Comments

No Known Activations