INDEX

Explanations

occurrences of the substring "eng" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

issy

-0.16

ogg

-0.14

 meanwhile

-0.14

TG

-0.14

 genuine

-0.14

uce

-0.14

 dims

-0.14

emploi

-0.13

 with

-0.13

bg

-0.13

POSITIVE LOGITS

yo

0.17

Ð°ÑĢÑĤÐ°Ð¼

0.15

asser

0.15

Ð¼Ð¾Ð²

0.15

doll

0.15

Ð°ÑĤÐ¾

0.15

slaught

0.14

inned

0.14

 lesser

0.14

(defvar

0.14

Activations Density 0.011%

occurrences of the substring "eng" in various contexts

No Comments

No Known Activations