INDEX

Explanations

punctuation marks, particularly periods and exclamation points

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

 rall

-0.81

 targ

-0.76

 volunte

-0.75

 uninterrupted

-0.73

 skelet

-0.72

perty

-0.72

 teasp

-0.72

 accomp

-0.72

 answ

-0.72

 behavi

-0.70

POSITIVE LOGITS

 Didn

1.03

 Exactly

1.02

 What

1.02

 Lots

1.00

 That

0.98

 Yeah

0.98

 Okay

0.97

 Maybe

0.96

 Something

0.96

 Anyway

0.96

Activations Density 0.046%

punctuation marks, particularly periods and exclamation points

No Comments

No Known Activations