INDEX

Explanations

references to microphones

oai_token-act-pair · gpt-3.5-turbo

mentions of microphones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Spit

-0.84

tenance

-0.82

ENGTH

-0.76

 Unified

-0.70

ACTED

-0.69

 Dull

-0.67

CLASSIFIED

-0.67

ignment

-0.67

ãģĦ

-0.67

 Pact

-0.65

POSITIVE LOGITS

roman

1.06

helle

1.00

ropolitan

0.99

romancer

0.96

rom

0.91

asonic

0.86

rics

0.85

rodu

0.84

rones

0.83

rop

0.82

Activations Density 0.011%

references to microphones

mentions of microphones

No Comments

No Known Activations