INDEX

Explanations

references to symbolic actions or gestures

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

ondo

-0.85

zona

-0.78

tics

-0.77

elsen

-0.76

 Nightmares

-0.73

raid

-0.72

ãĥ¯ãĥ³

-0.71

gres

-0.71

vae

-0.70

seed

-0.68

POSITIVE LOGITS

 gesture

1.44

 gestures

1.24

 toward

1.03

 towards

1.01

 greeting

0.90

 salute

0.81

 recognition

0.78

 dexterity

0.77

 waving

0.74

 indicating

0.72

Activations Density 0.008%

references to symbolic actions or gestures

No Comments

No Known Activations