INDEX

Explanations

adjectives related to quality or performance

oai_token-act-pair · gpt-3.5-turbo

terms related to mixed outcomes and varying quality

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

dfx

-0.73

 Encyclopedia

-0.72

 smashed

-0.61

 Ships

-0.61

RL

-0.61

oin

-0.61

enium

-0.61

Xan

-0.61

OM

-0.61

 Moons

-0.59

POSITIVE LOGITS

 distingu

0.72

ikawa

0.70

conserv

0.69

0.67

luster

0.67

 dism

0.67

abwe

0.65

 conduc

0.65

 fared

0.65

Compar

0.65

Activations Density 0.485%

adjectives related to quality or performance

terms related to mixed outcomes and varying quality

No Comments

No Known Activations