INDEX

Explanations

terms related to historical events, particularly focused on World War II

oai_token-act-pair · gpt-3.5-turbo

references to historical events and classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

IUM

-0.67

GUI

-0.66

 crates

-0.62

tnc

-0.61

CHAT

-0.61

dow

-0.61

 Manson

-0.59

MEN

-0.58

MSN

-0.58

 varied

-0.58

POSITIVE LOGITS

imester

1.04

endment

0.85

eve

0.80

amic

0.73

agonist

0.71

secut

0.70

arsity

0.68

milo

0.68

acia

0.67

roxy

0.65

Activations Density 0.294%

terms related to historical events, particularly focused on World War II

references to historical events and classifications

No Comments

No Known Activations