INDEX

Explanations

words related to persistence or continuous actions

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their actions or states in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 preferring

-0.68

 Variant

-0.67

 occurring

-0.65

 understatement

-0.64

 delaying

-0.64

 guiName

-0.62

Ø§ÙĦ

-0.61

actionDate

-0.61

Aim

-0.60

holding

-0.60

POSITIVE LOGITS

 reaches

1.10

 reach

0.98

 expires

0.97

 finally

0.95

 sufficiently

0.86

 exhaustion

0.85

 reached

0.85

 completes

0.84

 expire

0.83

 finishes

0.80

Activations Density 0.099%

words related to persistence or continuous actions

references to individuals and their actions or states in a narrative context

No Comments

No Known Activations