INDEX

Explanations

transitions in job roles or political affiliations

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

UAL

-0.70

clair

-0.68

Favorite

-0.68

CLUD

-0.64

caster

-0.64

 excel

-0.64

FIELD

-0.63

POST

-0.63

Monitor

-0.62

addock

-0.61

POSITIVE LOGITS

 replaced

0.94

 prematurely

0.93

 indefinitely

0.92

 altogether

0.91

 disgrace

0.87

 abruptly

0.85

:(

0.82

 resignation

0.82

 permanently

0.81

 expired

0.80

Activations Density 4.829%

transitions in job roles or political affiliations

No Comments

No Known Activations