INDEX

Explanations

phrases related to personal experiences or perspectives

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

ĸļ

-0.72

rants

-0.71

REDACTED

-0.69

bos

-0.67

pants

-0.66

 Olson

-0.66

zens

-0.66

escription

-0.65

iott

-0.64

sv

-0.64

POSITIVE LOGITS

 efficiency

1.20

 competitiveness

1.13

 productivity

1.09

 morale

1.06

 chances

1.03

 resilience

1.03

 reliability

0.98

 effectiveness

0.97

 hygiene

0.95

 responsiveness

0.95

Activations Density 0.183%

phrases related to personal experiences or perspectives

No Comments

No Known Activations