INDEX

Explanations

phrases related to interference or intervention in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to interference, particularly in political or social contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ndra

-0.81

sal

-0.75

chal

-0.74

":"/

-0.74

Bio

-0.72

True

-0.72

abc

-0.72

href

-0.71

Template

-0.69

mber

-0.68

POSITIVE LOGITS

 interference

1.11

 interfere

1.00

 interfering

0.98

 interfered

0.98

 meddling

0.95

 tampering

0.92

 medd

0.84

 newsp

0.78

 intr

0.76

 adversely

0.72

Activations Density 0.013%

phrases related to interference or intervention in various contexts

references to interference, particularly in political or social contexts

No Comments

No Known Activations