INDEX

Explanations

words related to recognition or acknowledgment of achievements and characteristics

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

hani

-0.15

rection

-0.15

andr

-0.15

iction

-0.14

 Campbell

-0.14

division

-0.14

/from

-0.14

áº¹n

-0.13

riage

-0.13

 Nath

-0.13

POSITIVE LOGITS

ablish

0.16

InThe

0.16

OrUpdate

0.15

entially

0.15

ÑģÑıÑĤ

0.15

lessly

0.15

incinn

0.15

ifies

0.15

atively

0.15

inize

0.15

Activations Density 0.130%

words related to recognition or acknowledgment of achievements and characteristics

No Comments

No Known Activations