INDEX

Explanations

references to military-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to military actions or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ì¶

-0.77

posure

-0.75

nder

-0.73

Charge

-0.69

Mist

-0.68

 Franken

-0.67

 Springs

-0.65

title

-0.64

missions

-0.64

PER

-0.64

POSITIVE LOGITS

 milit

1.29

 reluct

0.98

arily

0.94

ament

0.85

 guiActiveUn

0.84

ilitary

0.83

untled

0.83

 fatig

0.82

heast

0.76

 diseng

0.76

Activations Density 0.002%

references to military-related terms

terms related to military actions or organizations

No Comments

No Known Activations