INDEX

Explanations

statements related to hypothetical or speculative situations

oai_token-act-pair · gpt-3.5-turbo

conditional statements or hypotheses involving the word "would."

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

Kag

-0.67

rylic

-0.64

 Seoul

-0.63

building

-0.61

resources

-0.61

isted

-0.60

Pie

-0.58

meta

-0.58

 Corpus

-0.58

 Chal

-0.57

POSITIVE LOGITS

 gladly

0.93

be

0.92

 doubtless

0.87

 surely

0.87

 ideally

0.86

 suffice

0.84

 dearly

0.83

 otherwise

0.82

 undoubtedly

0.82

 classify

0.82

Activations Density 0.152%

statements related to hypothetical or speculative situations

conditional statements or hypotheses involving the word "would."

No Comments

No Known Activations