INDEX

Explanations

phrases related to legal cases or investigations

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Picks

-0.76

activity

-0.70

anza

-0.69

marine

-0.67

itivity

-0.65

 Catch

-0.65

itiveness

-0.64

ENSE

-0.64

hof

-0.64

TP

-0.64

POSITIVE LOGITS

 executed

1.21

 photographed

1.18

 repaired

1.16

 interrogated

1.15

 reused

1.14

 redistributed

1.12

 flown

1.12

 transported

1.12

 prosecuted

1.09

 tested

1.09

Activations Density 0.592%

phrases related to legal cases or investigations

No Comments

No Known Activations