INDEX

Explanations

affirmations or expressions of agreement

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

urger

-0.76

string

-0.67

rift

-0.66

minecraft

-0.66

tnc

-0.63

manufact

-0.63

division

-0.62

Lago

-0.62

graph

-0.62

due

-0.61

POSITIVE LOGITS

lahoma

1.16

bye

1.07

AY

1.02

Okay

0.98

Alright

0.88

terday

0.86

 okay

0.84

yeah

0.84

 yeah

0.83

OK

0.82

Activations Density 0.008%

affirmations or expressions of agreement

No Comments

No Known Activations