INDEX

Explanations

phrases expressing improvement or superiority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

phabet

-0.70

pex

-0.65

clerosis

-0.64

urses

-0.62

regate

-0.61

gemony

-0.61

iasm

-0.61

estones

-0.61

mad

-0.59

pson

-0.59

POSITIVE LOGITS

 than

1.12

 suited

1.08

than

0.81

served

0.80

 behaved

0.78

 situated

0.77

 Than

0.72

luaj

0.72

 safest

0.71

 served

0.67

Activations Density 0.082%

phrases expressing improvement or superiority

No Comments

No Known Activations