INDEX

Explanations

subtle elements or cues in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "subtle" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ETF

-0.72

|--

-0.65

 confidently

-0.65

 Journals

-0.65

Strong

-0.64

 largeDownload

-0.63

 CENT

-0.63

HOME

-0.61

arenthood

-0.61

TRI

-0.60

POSITIVE LOGITS

ty

0.98

 subtle

0.85

 Subtle

0.84

 nuances

0.80

aneous

0.75

 nuance

0.75

 asymm

0.74

asing

0.74

reet

0.74

aneously

0.74

Activations Density 0.009%

subtle elements or cues in various contexts

instances of the word "subtle" and its variations

No Comments

No Known Activations