INDEX

Explanations

repeated occurrences of the word "för" and related conjunctions

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

-0.17

cul

-0.15

pr

-0.15

 corrective

-0.14

ous

-0.14

rix

-0.14

elia

-0.14

rah

-0.14

rey

-0.14

POSITIVE LOGITS

ÑģÑĤÐµÑĢ

0.16

ilog

0.15

Ø¬ÛĮ

0.15

erif

0.15

ires

0.15

 Gratuit

0.14

.rf

0.14

 lá»¥c

0.14

chio

0.14

eneg

0.14

Activations Density 0.015%

repeated occurrences of the word "för" and related conjunctions

No Comments

No Known Activations