INDEX

Explanations

terms related to hostage situations

oai_token-act-pair · gpt-3.5-turbo

references to the term "hostage" and its various 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

aum

-0.72

iquette

-0.69

resso

-0.68

chlor

-0.66

odor

-0.63

appa

-0.63

unders

-0.63

orkshire

-0.62

orough

-0.61

ril

-0.60

POSITIVE LOGITS

 hostage

1.38

 hostages

1.20

 captives

0.97

 captive

0.94

 negotiator

0.88

 captivity

0.79

 kidn

0.79

 negotiators

0.77

 ransom

0.77

 rescuing

0.73

Activations Density 0.009%

terms related to hostage situations

references to the term "hostage" and its various contexts

No Comments

No Known Activations