INDEX

Explanations

terms related to horror or fearful situations

oai_token-act-pair · gpt-3.5-turbo

words related to horror or frightening experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

•layer 0 - fear

No Comments

Negative Logits

 Stat

-0.74

 Annotations

-0.65

 Prep

-0.64

 Commonwealth

-0.63

Cop

-0.61

 Qiao

-0.60

 Teacher

-0.59

 Statistics

-0.59

 Boost

-0.59

 Caller

-0.57

POSITIVE LOGITS

hor

4.75

Hor

2.24

hor

1.51

HOR

1.44

Hor

1.41

hal

1.32

ho

1.26

hum

1.24

hol

1.18

hov

1.17

Activations Density 0.011%

terms related to horror or fearful situations

words related to horror or frightening experiences

No Comments

No Known Activations