INDEX

Explanations

words related to high-ranking positions or leadership roles

oai_token-act-pair · gpt-3.5-turbo

references to leadership positions, particularly those of "head" or similar roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 IMAGES

-0.67

PsyNetMessage

-0.63

 subp

-0.62

Constructed

-0.61

 delta

-0.60

orum

-0.60

 mell

-0.59

Dup

-0.59

 autos

-0.58

POSITIVE LOGITS

quarter

1.15

lining

1.13

quartered

1.07

hun

1.05

gear

1.04

hunter

1.03

canon

1.03

liner

1.00

scar

0.98

quarters

0.98

Activations Density 0.025%

words related to high-ranking positions or leadership roles

references to leadership positions, particularly those of "head" or similar roles

No Comments

No Known Activations