INDEX

Explanations

combinations of letters and their occurrences in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

dum

-0.16

GES

-0.16

AYS

-0.15

nio

-0.15

cken

-0.14

(es

-0.14

orama

-0.14

imenti

-0.14

Ã©ri

-0.13

eko

-0.13

POSITIVE LOGITS

 Weak

0.16

 weak

0.15

odore

0.14

pton

0.14

ley

0.14

 Ting

0.14

rega

0.14

LocalizedString

0.13

xFF

0.13

Activations Density 0.186%

combinations of letters and their occurrences in different contexts

No Comments

No Known Activations