INDEX

Explanations

actions of attempting or trying to do something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "attempting" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Origin

-0.75

tal

-0.73

spl

-0.72

 Foot

-0.69

spr

-0.69

eye

-0.69

 Chop

-0.69

 Blessed

-0.66

 Wrap

-0.64

patch

-0.64

POSITIVE LOGITS

 carbohyd

0.97

pload

0.87

querque

0.87

CLASSIFIED

0.85

 unsuccessfully

0.85

Attempts

0.84

iculty

0.84

 guiActiveUn

0.84

tyr

0.82

 referen

0.81

Activations Density 0.009%

actions of attempting or trying to do something

instances of the word "attempting" in various contexts

No Comments

No Known Activations