INDEX

Explanations

proper noun phrases that could refer to historical events, items, or individuals

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

achus

-0.79

osate

-0.76

cation

-0.70

ob

-0.70

onis

-0.70

heit

-0.69

za

-0.67

posal

-0.67

Ã«

-0.67

fecture

-0.67

POSITIVE LOGITS

 kinds

1.13

 sorts

1.07

 facts

1.00

 types

0.95

 truths

0.94

 sentiments

0.91

 fellows

0.90

 thoughts

0.90

 qualities

0.88

 distinctions

0.87

Activations Density 0.754%

proper noun phrases that could refer to historical events, items, or individuals

No Comments

No Known Activations