INDEX

Explanations

phrases related to intentional or deliberate actions

oai_token-act-pair · gpt-3.5-turbo

terms related to intentional deceit or misleading actions

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

raph

-0.72

NOW

-0.72

masters

-0.68

soon

-0.67

 Warriors

-0.67

yip

-0.66

Liter

-0.66

 Warrant

-0.65

ival

-0.65

Legend

-0.65

POSITIVE LOGITS

 mislead

1.06

 sabot

1.03

 misled

1.01

 obfusc

1.01

 misrepresent

1.00

 misleading

0.99

 fals

0.98

 omitted

0.96

 omit

0.95

 dece

0.95

Activations Density 0.059%

phrases related to intentional or deliberate actions

terms related to intentional deceit or misleading actions

No Comments

No Known Activations