INDEX

Explanations

information about changes in job roles or responsibilities

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

raq

-0.81

heres

-0.73

Methods

-0.67

 Scores

-0.65

 Cruel

-0.65

lag

-0.64

rencies

-0.62

bidden

-0.62

ãĤµ

-0.61

apa

-0.61

POSITIVE LOGITS

 helm

1.30

 role

1.26

 roles

1.25

 duties

1.17

role

1.05

 overseeing

1.03

 slot

1.02

 mantle

1.01

 reins

1.00

 vacated

0.99

Activations Density 1.933%

information about changes in job roles or responsibilities

No Comments

No Known Activations