INDEX

Explanations

references to the number of employees in various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of employees in various contexts related to work

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Pac

-0.75

rous

-0.75

COMPLE

-0.71

soon

-0.68

Brow

-0.67

 Cause

-0.66

 horr

-0.66

bol

-0.64

forward

-0.64

las

-0.63

POSITIVE LOGITS

 Employees

0.82

arers

0.79

 employees

0.79

 cafeteria

0.78

hips

0.77

 personnel

0.76

onomy

0.73

 unions

0.73

 employee

0.71

 interns

0.69

Activations Density 0.035%

references to the number of employees in various contexts

mentions of employees in various contexts related to work

No Comments

No Known Activations