INDEX

Explanations

elements related to processes and procedures in various contexts

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

ãĤ©

-0.71

igl

-0.71

ĪĴ

-0.69

 bonded

-0.65

çīĪ

-0.65

çĶ°

-0.62

 glim

-0.62

bern

-0.61

 nutshell

-0.61

ificent

-0.61

POSITIVE LOGITS

 altogether

1.38

 indefinitely

1.14

 lest

1.07

 prematurely

0.90

 because

0.90

due

0.89

 entirely

0.88

 until

0.88

 anymore

0.88

until

0.86

Activations Density 0.187%

elements related to processes and procedures in various contexts

No Comments

No Known Activations