INDEX

Explanations

semicolons representing different actions or activities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

din

-0.77

orts

-0.76

stration

-0.73

inki

-0.72

ractions

-0.72

ħĭ

-0.71

ront

-0.71

ricular

-0.70

ookie

-0.69

hedral

-0.69

POSITIVE LOGITS

-)

1.13

 furthermore

1.06

 alternatively

1.04

 moreover

0.99

âĢ¢âĢ¢âĢ¢âĢ¢

0.96

 consequently

0.90

 thence

0.89

 hence

0.89

 namely

0.88

 meanwhile

0.87

Activations Density 0.716%

semicolons representing different actions or activities

No Comments

No Known Activations