INDEX

Explanations

names of places, organizations, and locations

oai_token-act-pair · gpt-3.5-turbo

references to violence and conflict-related events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

advertisement

-0.94

ADVERTISEMENT

-0.90

âĢķ

-0.86

âĢ¦âĢ¦âĢ¦âĢ¦

-0.84

 tremend

-0.84

,)

-0.83

CLASSIFIED

-0.82

—-

-0.81

 princ

-0.77

:(

-0.77

POSITIVE LOGITS

 Nato

1.41

BBC

1.24

BBC

1.19

Image

1.18

 (Â£

1.10

 apologised

1.06

 Nasa

1.06

Analysis

1.04

 recognise

0.94

 criticised

0.94

Activations Density 0.265%

names of places, organizations, and locations

references to violence and conflict-related events

No Comments

No Known Activations