INDEX

Explanations

phrases related to government officials and their actions

oai_token-act-pair · gpt-3.5-turbo

references to authorities or officials

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.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

oise

-0.70

OPE

-0.68

ãĤ¦

-0.67

ocene

-0.67

esville

-0.64

oons

-0.63

Interstitial

-0.63

Beg

-0.61

ãĥ¤

-0.60

[|

-0.60

POSITIVE LOGITS

 overseeing

1.10

 briefed

1.05

 tasked

0.98

doms

0.94

 responsible

0.89

 familiar

0.87

 stationed

0.86

 involved

0.81

 complicit

0.79

 investigating

0.79

Activations Density 0.066%

phrases related to government officials and their actions

references to authorities or officials

No Comments

No Known Activations