INDEX

Explanations

mentions of future actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "will" indicating future events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

reality

-0.65

 abstraction

-0.62

amped

-0.62

Lear

-0.62

Mom

-0.61

ZI

-0.60

processing

-0.60

-0.59

establishment

-0.59

engineering

-0.59

POSITIVE LOGITS

be

1.19

 continue

1.05

 undoubtedly

1.05

 doubtless

1.04

 likely

1.01

 surely

0.97

 gladly

0.95

 probably

0.93

 remain

0.93

 definitely

0.92

Activations Density 0.205%

mentions of future actions

instances of the word "will" indicating future events or actions

No Comments

No Known Activations