INDEX

Explanations

words related to criticism and negative consequences

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

inarily

-0.69

craft

-0.66

icipated

-0.64

riots

-0.64

issance

-0.64

-0.63

hyde

-0.62

ords

-0.62

igi

-0.60

POSITIVE LOGITS

 much

0.89

 busy

0.84

far

0.83

 tempting

0.79

 risky

0.79

 many

0.77

 afraid

0.77

oths

0.76

 distracting

0.74

ls

0.72

Activations Density 0.326%

words related to criticism and negative consequences

No Comments

No Known Activations