INDEX

Explanations

references to specific names and characters in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

æĤł

-0.15

 Jail

-0.14

,readonly

-0.14

stile

-0.14

reverse

-0.14

finity

-0.14

[section

-0.14

ddit

-0.14

 pimp

-0.13

æķ¦

-0.13

POSITIVE LOGITS

leton

0.17

onia

0.17

deen

0.15

åį

0.15

pedo

0.15

Ð»Ð¾Ð¶

0.15

inct

0.14

isible

0.14

ÑİÐ½

0.14

ierge

0.14

Activations Density 0.047%

references to specific names and characters in a narrative context

No Comments

No Known Activations