INDEX

Explanations

references to specific media titles and names related to entertainment and movies

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

agara

-0.90

bsite

-0.85

itary

-0.83

ordering

-0.82

iple

-0.82

elled

-0.82

abase

-0.81

medium

-0.80

orate

-0.80

Äĩ

-0.79

POSITIVE LOGITS

 Reloaded

1.16

 Cowboy

1.15

 Madness

1.15

 Strategies

1.12

 Dreams

1.10

 Express

1.09

 Solutions

1.08

 Mania

1.08

 Nation

1.07

 Nights

1.07

Activations Density 0.095%

references to specific media titles and names related to entertainment and movies

No Comments

No Known Activations