INDEX

Explanations

legal terms and concepts

oai_token-act-pair · gpt-3.5-turbo

keywords indicating significant entities or concepts related to law and social dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

IVERS

-0.69

details

-0.67

 Valhalla

-0.67

ãĤ¼ãĤ¦ãĤ¹

-0.64

 Facilities

-0.63

yrics

-0.62

BG

-0.61

nance

-0.59

une

-0.59

DEF

-0.59

POSITIVE LOGITS

 apiece

1.07

per

0.86

undred

0.74

 alone

0.71

stroke

0.71

 each

0.71

secut

0.69

 parcel

0.69

ilion

0.67

abyte

0.67

Activations Density 0.233%

legal terms and concepts

keywords indicating significant entities or concepts related to law and social dynamics

No Comments

No Known Activations