INDEX

Explanations

mentions of stun guns and stun-related activities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stun" and its variations

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

Not in Any Lists

No Comments

Negative Logits

gio

-0.80

Catal

-0.76

 Sabb

-0.75

abb

-0.74

versions

-0.71

isl

-0.71

adata

-0.71

isle

-0.67

lique

-0.66

Comm

-0.66

POSITIVE LOGITS

 stun

3.55

 Stun

2.30

 stunned

1.57

 incapac

1.33

 cooldown

1.14

 silenced

1.13

 choke

1.13

 stim

1.05

 immobil

1.03

 taunt

1.02

Activations Density 0.038%

mentions of stun guns and stun-related activities

occurrences of the word "stun" and its variations

No Comments

No Known Activations