INDEX

Explanations

names of people or entities in the tech or academic field

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

Tokens

-0.80

 respectively

-0.73

Compare

-0.71

$.

-0.71

depending

-0.70

cause

-0.67

Xi

-0.64

SpaceEngineers

-0.62

ADRA

-0.61

adra

-0.61

POSITIVE LOGITS

 writes

0.97

(@

0.93

 wrote

0.92

 Photography

0.92

 Profile

0.89

 celebrates

0.88

 remembers

0.88

 teaches

0.87

 greets

0.84

Jr

0.82

Activations Density 0.309%

names of people or entities in the tech or academic field

No Comments

No Known Activations