INDEX

Explanations

words related to radio technology

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Catalyst

-0.69

arez

-0.68

gio

-0.68

 ANGEL

-0.67

gil

-0.66

 Kumar

-0.64

gur

-0.64

ortium

-0.63

adra

-0.63

AGES

-0.62

POSITIVE LOGITS

actively

1.01

active

0.94

activity

0.88

 wagon

0.86

oute

0.84

 stations

0.83

radio

0.82

 station

0.77

 radio

0.76

oded

0.76

Activations Density 1.174%

words related to radio technology

No Comments

No Known Activations