INDEX

Explanations

phrases that denote opportunities or possibilities for action

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

atoon

-0.75

ileaks

-0.69

 Logged

-0.65

Zip

-0.64

indle

-0.63

 spills

-0.62

 Appalach

-0.61

 attm

-0.59

iba

-0.59

irit

-0.58

POSITIVE LOGITS

 choice

0.97

 opportunity

0.89

 permission

0.88

 chance

0.85

choice

0.79

treatment

0.77

 ration

0.75

 licence

0.72

otomy

0.72

 latitude

0.71

Activations Density 0.070%

phrases that denote opportunities or possibilities for action

No Comments

No Known Activations