INDEX

Explanations

terms related to primitive concepts or states

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

ojÃŃ

-0.16

rg

-0.15

agged

-0.15

ampa

-0.15

oth

-0.14

scaling

-0.14

íĥĿ

-0.14

DOM

-0.13

iÄĩ

-0.13

esco

-0.13

POSITIVE LOGITS

SPATH

0.15

PARATOR

0.15

Rocket

0.14

swick

0.14

imon

0.14

/native

0.14

onds

0.14

USTER

0.13

SSF

0.13

Activations Density 0.009%

terms related to primitive concepts or states

No Comments

No Known Activations