INDEX

Explanations

names of individuals associated with controversial topics or issues

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

iP

-0.86

 RELE

-0.84

IMP

-0.80

 INTER

-0.77

 acron

-0.77

ASC

-0.76

 supers

-0.74

 DIRECT

-0.71

SAN

-0.71

 CONTR

-0.69

POSITIVE LOGITS

inar

1.12

idy

1.10

aga

1.09

axis

1.05

ady

1.00

ipal

1.00

ucket

1.00

arie

0.98

ilus

0.97

actor

0.97

Activations Density 2.075%

names of individuals associated with controversial topics or issues

No Comments

No Known Activations