INDEX

Explanations

phrases related to reflection or representation

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

queue

-0.78

 contend

-0.77

sites

-0.72

jan

-0.70

 headlined

-0.66

efer

-0.65

opher

-0.64

jet

-0.64

BILL

-0.63

parse

-0.63

POSITIVE LOGITS

ively

0.92

ational

0.86

 sentiments

0.83

 eternity

0.81

orical

0.80

orically

0.76

iveness

0.74

matically

0.74

atively

0.74

 purity

0.73

Activations Density 0.864%

phrases related to reflection or representation

No Comments

No Known Activations