INDEX

Explanations

proper nouns related to political figures or entities

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

estone

-0.77

lez

-0.73

oor

-0.73

ords

-0.72

estones

-0.70

away

-0.67

abouts

-0.67

zl

-0.66

stone

-0.66

oÄŁ

-0.66

POSITIVE LOGITS

 Mike

0.99

 Vincent

0.93

Jim

0.92

 Denis

0.91

 Colin

0.91

 Steve

0.90

 Philip

0.89

Sue

0.88

 Gerard

0.88

 Leslie

0.88

Activations Density 0.111%

proper nouns related to political figures or entities

No Comments

No Known Activations