INDEX

Explanations

expressions of desire or intention

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

VERTISEMENT

-0.78

rir

-0.66

ulty

-0.63

semble

-0.62

icol

-0.62

trust

-0.60

RL

-0.59

anka

-0.58

cession

-0.58

NVIDIA

-0.57

POSITIVE LOGITS

 revenge

0.94

reprene

0.86

lessly

0.83

to

0.83

 clarification

0.78

 attention

0.77

 answers

0.75

 assurances

0.72

 desperately

0.70

 somebody

0.69

Activations Density 2.553%

expressions of desire or intention

No Comments

No Known Activations