INDEX

Explanations

sentences describing step-by-step processes or instructions

oai_token-act-pair · gpt-3.5-turbo

actionable items or strategies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Another

-0.65

Another

-0.61

 anymore

-0.59

 realise

-0.57

another

-0.57

official

-0.57

Nevertheless

-0.57

gotten

-0.53

Still

-0.53

ibur

-0.53

POSITIVE LOGITS

 ranging

0.93

 grouped

0.88

 varying

0.87

 each

0.86

 preferably

0.84

 ranges

0.80

 categorized

0.80

 totaling

0.79

 progressively

0.78

 respectively

0.76

Activations Density 1.057%

sentences describing step-by-step processes or instructions

actionable items or strategies

No Comments

No Known Activations