INDEX

Explanations

words related to military or police activities

oai_token-act-pair · gpt-3.5-turbo

mentions of specific operations or activities, particularly in a logistical or strategic context

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

cknowled

-0.70

bitious

-0.70

hner

-0.65

gets

-0.64

ixed

-0.64

cil

-0.64

cit

-0.63

evil

-0.63

attering

-0.63

ako

-0.62

POSITIVE LOGITS

 operation

1.06

eering

0.97

eers

0.87

 operations

0.84

operation

0.80

quit

0.79

 operative

0.76

 Werewolf

0.73

 operator

0.73

 Operation

0.73

Activations Density 0.017%

words related to military or police activities

mentions of specific operations or activities, particularly in a logistical or strategic context

No Comments

No Known Activations