INDEX

Explanations

passive or negative terms related to consequences or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to passive strategies and effects in complex systems

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

Cash

-0.66

Doctors

-0.65

Tec

-0.62

Rap

-0.62

 Chap

-0.60

Stars

-0.54

roma

-0.54

Benz

-0.54

 Â®

-0.54

Mex

-0.53

POSITIVE LOGITS

leground

0.63

 refere

0.62

oser

0.61

ness

0.61

OVA

0.57

ailability

0.57

ir

0.57

ability

0.57

wart

0.57

unction

0.55

Activations Density 1.058%

passive or negative terms related to consequences or actions

terms related to passive strategies and effects in complex systems

No Comments

No Known Activations