INDEX

Explanations

phrases related to specific events or activities happening during a particular time period

oai_token-act-pair · gpt-3.5-turbo

references to events or processes occurring during specific timeframes

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

osponsors

-0.65

ographically

-0.62

facts

-0.62

fortune

-0.62

ById

-0.61

coins

-0.58

 reused

-0.58

 killer

-0.58

prints

-0.57

detail

-0.57

POSITIVE LOGITS

 period

1.21

 periods

1.13

 lull

1.02

 hiatus

0.99

imester

0.96

 interval

0.93

era

0.93

 downtime

0.92

period

0.92

 recess

0.91

Activations Density 0.264%

phrases related to specific events or activities happening during a particular time period

references to events or processes occurring during specific timeframes

No Comments

No Known Activations