INDEX

Explanations

adjectives describing opinion or evaluation

oai_token-act-pair · gpt-3.5-turbo

assertions and tentative claims in statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

doors

-0.66

 PLAN

-0.61

DO

-0.57

 unfocusedRange

-0.57

SN

-0.57

 Everywhere

-0.56

 deadlines

-0.55

 manual

-0.54

 freezes

-0.53

XY

-0.53

POSITIVE LOGITS

 understatement

0.83

 compare

0.82

 conjecture

0.79

rued

0.79

 recall

0.76

 comparison

0.75

 exaggeration

0.75

athom

0.75

haps

0.75

 Compare

0.74

Activations Density 0.831%

adjectives describing opinion or evaluation

assertions and tentative claims in statements

No Comments

No Known Activations