INDEX

Explanations

references to historical events or information

oai_token-act-pair · gpt-3.5-turbo

references to historical events or contexts

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

nery

-0.88

lain

-0.85

Downloadha

-0.82

geon

-0.81

ertodd

-0.80

Flow

-0.78

DAQ

-0.76

aye

-0.74

rosis

-0.72

Kids

-0.71

POSITIVE LOGITS

orical

1.04

orically

1.00

 preservation

0.96

 significance

0.93

 revision

0.89

 precedent

0.85

 relics

0.84

 lows

0.82

 Context

0.80

 inaccur

0.78

Activations Density 0.044%

references to historical events or information

references to historical events or contexts

No Comments

No Known Activations