INDEX

Explanations

assertive statements or declarations that are straightforward or explicit

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

eries

-0.72

 resil

-0.71

 passively

-0.69

 Loft

-0.68

INTER

-0.67

 tremend

-0.66

ITAL

-0.65

nesota

-0.64

inse

-0.64

rology

-0.64

POSITIVE LOGITS

ances

1.16

cut

1.05

iary

0.95

 deline

0.93

ance

0.92

 indication

0.82

headed

0.82

 distinction

0.81

ively

0.79

faced

0.77

Activations Density 1.732%

assertive statements or declarations that are straightforward or explicit

No Comments

No Known Activations