INDEX

Explanations

instances of conjunctions and references to percentages or numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

rients

-0.16

ultan

-0.15

 Giang

-0.15

erus

-0.14

fcc

-0.14

(KP

-0.14

èĪ¬

-0.14

ddit

-0.13

rapy

-0.13

 báºŃc

-0.13

POSITIVE LOGITS

emi

0.14

shi

0.14

arem

0.14

 Filed

0.14

 previously

0.14

 Cons

0.14

Ã¤d

0.14

leg

0.14

aren

0.14

 recht

0.14

Activations Density 0.074%

instances of conjunctions and references to percentages or numerical data

No Comments

No Known Activations